BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Prof Robert Griffiths (Monash University)
DTSTART:20200430T010000Z
DTEND:20200430T020000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/1
DESCRIPTION:Title: Lambda coalescent trees and graphs\nby Prof Rob
ert Griffiths (Monash University) as part of La Trobe University Statistic
s and Stochastic zoom seminar\n\n\nAbstract\nThe Lambda coalescent introdu
ced by Pitman (1999) and Sagitov (1999) is a random tree which has multip
le mergers. It is a dual to a Lambda-Fleming-Viot process which describes
a population of individuals with births and deaths\, where a single indiv
idual's children can contribute a large proportion of the population. The
population process has jumps at times where individuals give birth. The
Wright-Fisher diffusion in contrast\, being a diffusion\, is continuous o
ver time. The Kingman coalescent\, a random binary tree\, is dual to the
Wright-Fisher diffusion.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
1/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr Jessica Kasza (Monash University)
DTSTART:20200507T020000Z
DTEND:20200507T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/2
DESCRIPTION:Title: Cross-overs\, stepped wedges and staircases: some r
ecent work in longitudinal cluster randomised trials\nby Dr Jessica Ka
sza (Monash University) as part of La Trobe University Statistics and Stoc
hastic zoom seminar\n\n\nAbstract\nAlthough individually randomised trials
are the gold standard for assessing the impact of new treatments on pati
ent outcomes\, cluster randomised trials are necessary when testing the e
ffect of healthcare provider-level changes on patient outcomes\, e.g. the
effect of a hospital-wide handwashing program on the number of patients
who acquire infections in hospital. Cluster randomised trials often requi
re large numbers of clusters and thus can be infeasible\, but longitudinal
cluster randomised trials\, where clusters may switch between interventio
n and control\, require smaller sample sizes. Cross-overs\, stepped wedge
s and staircases are all particular variants of longitudinal cluster rand
omised trials that are being conducted with increasing frequency. However
\, many of the underlying statistical aspects of these designs remain und
er-explored.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
2/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr David Frazier (Monash University)
DTSTART:20200611T020000Z
DTEND:20200611T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/3
DESCRIPTION:Title: Robust and Efficient Approximate Bayesian Computati
on: A Minimum Distance Approach\nby Dr David Frazier (Monash Universit
y) as part of La Trobe University Statistics and Stochastic zoom seminar\n
\n\nAbstract\nIn many instances\, the application of approximate Bayesian
methods is hampered by two practical features: 1) the requirement to proje
ct the data down to low-dimensional summary\, including the choice of this
projection\, and which ultimately yields inefficient inference\; 2) a pos
sible lack of robustness of these methods to deviations from the underlyin
g model structure. Motivated by these efficiency and robustness concerns\,
we construct a new Bayesian method that can deliver efficient estimators
when the underlying model is well-specified\, and which is simultaneously
robust to certain forms of model misspecification. This new approach bypas
ses the calculation of summaries by considering a norm between empirical a
nd simulated probability measures. For specific choices of the norm\, we d
emonstrate that this approach can be as efficient as exact Bayesian infere
nce\, and is robust to deviations from the underlying model assumptions. W
e illustrate this approach using several examples that have featured in th
e literature.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
3/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr Damjan Vukcevic (University of Melbourne)
DTSTART:20200618T020000Z
DTEND:20200618T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/4
DESCRIPTION:Title: Analysis of repeated categorical ratings: going bey
ond inter-rater agreement.\nby Dr Damjan Vukcevic (University of Melbo
urne) as part of La Trobe University Statistics and Stochastic zoom semina
r\n\n\nAbstract\nA common task in health and medicine is the classificatio
n of patient information into one of several categories by a trained exper
t. This could include assessing the presence and type of a tumour from a m
edical image or providing a disease diagnosis from a series of medical tes
ts. Often such judgements are hard to make and error prone: two experts ma
y rate the same scenario differently or the same expert may provide altern
ative ratings of the same scenario when rating it multiple times on differ
ent occasions.\n\nAnalysing the performance of such expert ‘raters’\,
and the accuracy of their ‘ratings’ across a series of ‘items’\, i
s a common theme in much of the health and medical literature\, especially
in the setting where the true underlying category is unknown. Existing ap
proaches\, such as Cohen’s kappa\, focus only on assessing inter-agreeme
nt\, and have known problems stemming from the lack of any notion of under
lying truth and the difficulty of coping with repeated ratings by the same
rater.\n\nHere we present and implement methods that explicitly model an
underlying true category for each item and can cope naturally with any num
ber of ratings for each item\, including repeated ratings by the same rate
r. We implement Bayesian versions of these models using the probabilistic
programming language Stan\, and create an R package to fit and interrogate
the output of these models.\n\nUsing real and simulated datasets\, which
are designed to mimic a wide range of medical scenarios\, we test the perf
ormance of these models in estimating the true class of each item. We also
explore situations such as having raters with much poorer accuracy\, and
comparisons with other (non-model-based) approaches.\n\nA common task in h
ealth and medicine is the classification of patient information into one o
f several categories by a trained expert. This could include assessing the
presence and type of a tumour from a medical image or providing a disease
diagnosis from a series of medical tests. Often such judgements are hard
to make and error prone: two experts may rate the same scenario differentl
y or the same expert may provide alternative ratings of the same scenario
when rating it multiple times on different occasions.\n\nAnalysing the per
formance of such expert ‘raters’\, and the accuracy of their ‘rating
s’ across a series of ‘items’\, is a common theme in much of the hea
lth and medical literature\, especially in the setting where the true unde
rlying category is unknown. Existing approaches\, such as Cohen’s kappa\
, focus only on assessing inter-agreement\, and have known problems stemmi
ng from the lack of any notion of underlying truth and the difficulty of c
oping with repeated ratings by the same rater.\n\nHere we present and impl
ement methods that explicitly model an underlying true category for each i
tem and can cope naturally with any number of ratings for each item\, incl
uding repeated ratings by the same rater. We implement Bayesian versions o
f these models using the probabilistic programming language Stan\, and cre
ate an R package to fit and interrogate the output of these models.\n\nUsi
ng real and simulated datasets\, which are designed to mimic a wide range
of medical scenarios\, we test the performance of these models in estimati
ng the true class of each item. We also explore situations such as having
raters with much poorer accuracy\, and comparisons with other (non-model-b
ased) approaches.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
4/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr Mumtaz Hussain (La Trobe University)
DTSTART:20200625T020000Z
DTEND:20200625T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/5
DESCRIPTION:Title: Metric number theory via geometry and dynamics: Mah
ler to Margulis\nby Dr Mumtaz Hussain (La Trobe University) as part of
La Trobe University Statistics and Stochastic zoom seminar\n\n\nAbstract\
nThere are two well-known approaches in solving the measure theoretic prob
lems in Diophantine approximation. The metrical approach arise from the g
eometry of numbers and the ergodic theoretic approach arise from the dynam
ics on the space of lattices. One of the main ingredients in the geometry
of numbers is the usage of Borel-Cantelli lemmas from probability theory.
Dynamics on the space of lattices rely on the Dani correspondence principl
e (1985) which was extensively developed further by Margulis and Kleinboc
k. I will discuss both of these approaches and along the way discuss some
well-known results such as the resolutions of Oppenheim (1929)\, Mahler (
1932) and Sprindzuk (1965) conjectures which influenced my research in th
e last few years.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
5/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Marina Masioti (La Trobe University)
DTSTART:20200603T010000Z
DTEND:20200603T020000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/6
DESCRIPTION:Title: Optimal transformations for dimension reduction and
the problem of eigenvalue switching\nby Marina Masioti (La Trobe Univ
ersity) as part of La Trobe University Statistics and Stochastic zoom semi
nar\n\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
6/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Jason Gavin Grealey (La Trobe University)
DTSTART:20200319T040000Z
DTEND:20200319T050000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/7
DESCRIPTION:Title: Investigating the utility of neural networks in gen
omic prediction\nby Jason Gavin Grealey (La Trobe University) as part
of La Trobe University Statistics and Stochastic zoom seminar\n\nAbstract:
TBA\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
7/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Illia Donhauzer (La Trobe University)
DTSTART:20200904T020000Z
DTEND:20200904T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/8
DESCRIPTION:Title: Asymptotic behaviour of functionals of random field
s\nby Illia Donhauzer (La Trobe University) as part of La Trobe Univer
sity Statistics and Stochastic zoom seminar\n\n\nAbstract\nThe talk is abo
ut the asymptotic behaviour of functionals of long-range dependent random
fields. The Strong Law of Large Numbers (SLLN) and new properties of the l
imit processes in the Non-central Limit Theorem (NLT) will be discussed.\n
\nThe SLLN for integral functionals of random fields with unboundedly incr
easing covariances will be presented. The SLLN is derived for the case of
increasing domains. Conditions on covariance functions such that the SLLN
holds will be provided. The considered scenarios include non-stationary ra
ndom fields. The discussion about applications to weak and long-range depe
ndent random fields and numerical examples will be shown.\n\nNew propertie
s of generalized Hermite-type processes that arise in NLT for integral fun
ctionals of long-range dependent random fields will be demonstrated. Contr
ary to the classical one-dimensional case\, it will be shown that for any
choice of a multidimensional observation window the generalized Hermite-ty
pe process has non-stationary increments.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
8/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr Jie Yen Fan (Monash University)
DTSTART:20201001T020000Z
DTEND:20201001T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/9
DESCRIPTION:Title: Multi-type age-structured population model\nby
Dr Jie Yen Fan (Monash University) as part of La Trobe University Statisti
cs and Stochastic zoom seminar\n\n\nAbstract\nPopulation process in genera
l setting\, where each individual reproduces and dies depending on the sta
te (such as age and type) of the individual as well as the entire populati
on\, offers a more realistic framework to population modelling. Formulatin
g the population dynamics as a measure-valued stochastic process allows us
to incorporate such dependence. We describe the dynamics of a multi-type
age-structured population as a measure-valued process\, and obtain its asy
mptotics\, in particular\, the law of large numbers and the central limit
theorem.\n\nJoint work with Kais Hamza\, Peter Jagers and Fima Klebaner.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
9/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Mason Terrett (La Trobe University)
DTSTART:20201001T030500Z
DTEND:20201001T040000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/10
DESCRIPTION:Title: SARGDV: Efficient identification of groundwater-de
pendent vegetation using synthetic aperture radar\nby Mason Terrett (L
a Trobe University) as part of La Trobe University Statistics and Stochast
ic zoom seminar\n\n\nAbstract\nGroundwater depletion impacts the sustainab
ility of numerous groundwater-dependent vegetation (GDV) globally\, placin
g significant stress on their capacity to provide environmental and ecolog
ical support for flora\, fauna\, and anthropic benefits. Cost effective me
thods of GDV identification will enable strategic protection of these crit
ical ecological systems\, through improved and sustainable groundwater man
agement by communities and industry. Recent application of synthetic apert
ure radar (SAR) earth observation data in Australia has demonstrated the u
tility of radar for identifying terrestrial groundwater-dependent ecosyste
ms at scale. Our research included the development of SARGDV\, a binary cl
assification model\, which uses the extreme gradient boosting (XGBoost) al
gorithm in conjunction with three data cubes composed of Sentinel-1 SAR in
terferometric wide images. Our method may be used to support the protectio
n of GDV communities globally by providing a long term\, cost-effective so
lution to identify GDVs over variable regions and climates\, via the use o
f freely available\, high-resolution\, globally available Sentinel-1 SAR d
ata sets. Our method offers global water management agencies a means towar
d more sustainable management of regional groundwater resources by providi
ng an efficient method to identify significant GDV occurrence within areas
where substantial groundwater extraction is ongoing.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
10/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr Tingjin Chu (University of Melbourne)
DTSTART:20201015T010000Z
DTEND:20201015T020000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/11
DESCRIPTION:Title: Large Spatial Data Modeling and Analysis: A Krylov
Subspace Approach\nby Dr Tingjin Chu (University of Melbourne) as par
t of La Trobe University Statistics and Stochastic zoom seminar\n\n\nAbstr
act\nEstimating the parameters of spatial models for large spatial dataset
s can be computationally challenging\, as it involves repeated evaluation
of sizable spatial covariance matrices. In this paper\, we aim to develop
Krylov subspace based methods that are computationally efficient for large
spatial data. Specifically\, we approximate the inverse and the log-deter
minant of the spatial covariance matrix in the log-likelihood function via
conjugate gradient and stochastic Lanczos on a Krylov subspace. These met
hods reduce the computational complexity from $O(N^3)$ to $O(N^2)$ and $O(
N\\log N)$ for dense and sparse matrices\, respectively. Moreover\, we qua
ntify the difference between the approximated log-likelihood function and
the original log-likelihood function and establish the consistency of para
meter estimates. Simulation studies are conducted to examine the computat
ional efficiency as well as the finite-sample properties. For illustration
\, our methodology is applied to analyze a large LiDAR dataset.\n\nThis is
joint work with Jialuo Liu\, Jun Zhu and Haonan Wang.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
11/
END:VEVENT
BEGIN:VEVENT
SUMMARY:4 presenters (La Trobe University)
DTSTART:20201105T230000Z
DTEND:20201106T010000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/12
DESCRIPTION:Title: Theses talks by 4 students\nby 4 presenters (L
a Trobe University) as part of La Trobe University Statistics and Stochast
ic zoom seminar\n\n\nAbstract\nPresented by:\n\n10.00am Vibhooti Bhatnagar
. A comparison of AIC and nested t-tests for nested model selection.\n\n10
.25am Navdeep Kaur. Is corrected AIC really better than AIC?\n\n10.50am Sa
tbir Kaur Bansal. Visualization of Variability of AIC.\n\n11.15 Ravindra N
ath Dahal. A review of Prediction Intervals obtained from model free machi
ne learning algorithms for point prediction\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
12/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof Gael M. Martin (Monash University)
DTSTART:20201118T233000Z
DTEND:20201119T003000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/13
DESCRIPTION:Title: Computing Bayes: Bayesian Computation from 1763 to
the 21st Century\nby Prof Gael M. Martin (Monash University) as part
of La Trobe University Statistics and Stochastic zoom seminar\n\n\nAbstrac
t\nThe Bayesian statistical paradigm uses the language of probability to e
xpress uncertainty about the phenomena that generate observed data. Probab
ility distributions thus characterize Bayesian inference\, with the rules
of probability used to transform prior probability distributions for all u
nknowns - models\, parameters\, latent variables - into posterior distribu
tions\, subsequent to the observation of data. Conducting Bayesian inferen
ce requires the evaluation of integrals in which these probability distrib
utions appear. Bayesian computation is all about evaluating such integrals
in the typical case where no analytical solution exists. This paper takes
the reader on a chronological tour of Bayesian computation over the past
two and a half centuries. Beginning with the one-dimensional integral firs
t confronted by Bayes in 1763\, through to recent problems in which the un
knowns number in the millions\, we place all computational problems into a
common framework\, and describe all computational methods using a common
notation. The aim is to help new researchers in particular - and more gene
rally those interested in adopting a Bayesian approach to empirical work -
make sense of the plethora of computational techniques that are now on of
fer\; understand when and why different methods are useful\; and see the l
inks that do exist\, between them all.\n\nJoint results with David T. Fraz
ier (Monash University) and Christian P. Robert (University of Dauphine\,
Paris). The paper appears as an arXiv pre-print. We are revising it at the
moment\, but it won't change in its essence: https://arxiv.org/pdf/2004.0
6425.pdf\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
13/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Ravindi Nanayakkara (La Trobe University)
DTSTART:20201119T030000Z
DTEND:20201119T040000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/14
DESCRIPTION:Title: Stochastic Modelling and Statistical Analysis of D
ependent Data\nby Ravindi Nanayakkara (La Trobe University) as part of
La Trobe University Statistics and Stochastic zoom seminar\n\n\nAbstract\
nFirst\, we discuss the obtained results about the analysis of spherical m
onofractal and multifractal random fields with cosmological applications.
The Rényi function plays an important role in the analysis of multifracta
l random fields. For random fields on the sphere\, there are three models
in the literature where the Rényi function is known explicitly [1]. The m
ain statistical model used to describe CMB data in the literature is isotr
opic Gaussian fields. We present some new theoretical models\, numerical m
ultifractality studies and methodology based on simulating random fields\,
computing the Rényi function and the multifractal spectrum for different
scenarios and actual CMB data. The results suggest that there may exist a
very minor multifractality of the CMB data [2].\n\nNext\, we discuss the
obtained results about the asymptotic normality of simultaneous estimators
of cyclic long-memory processes. Spectral singularities at non-zero frequ
encies play an important role in investigating cyclic or seasonal time ser
ies. The publication [3] introduced the generalized filtered method-of-mom
ents approach to simultaneously estimate singularity location and long-mem
ory parameters. This study [4] continues investigations of these simultane
ous estimators. The results about asymptotic normality of several statisti
cs are obtained. The methodology includes wavelet transformations as a par
ticular case. The theoretical findings are illustrated by numerical result
s including Meyer\, Shannon father wavelets and Mexican hat wavelets.\n\nF
inally\, we discuss multifractionality of spherical random fields with cos
mological applications. The Hölder exponent is used to measure the roughn
ess in a rigorous mathematical way [5]. In this study\, one dimensional an
d two dimensional pointwise Hölder exponent values are computed for the C
MB data using the HEALPix ring ordering and nested ordering visualisations
. The results suggest that there exist a considerable multifractionality i
n CMB data.\n\nReferences:\n\n Leonenko\, N. & Shieh\, N.R. (2013). Ré
nyi function for multifractal random fields. Fractals\, 21(2)\, 1350009.\n
Leonenko\, N.\, Nanayakkara\, R.\, & Olenko\, A. (2020). Analysis of S
pherical Monofractal and Multifractal Random Fields. Stochastic Environmen
tal Research and Risk Assessment Journal. https://doi.org/10.1007/s00477-0
20-01911-z\n Alomari\, H. M.\, Ayache\, A.\, Fradon\, M. & Olenko\, A.
(2020). Estimation of cyclic long-memory parameters. Scandinavian Journal
of Statistics\, 47(1) 104-133.\n Ayache\, A.\, Fradon\, M.\, Nanayakkar
a\, R.\, & Olenko\, A. (2020). Asymptotic normality of simultaneous estima
tors of cyclic long-memory processes. Submitted.\n Ayache\, A.\, & Véh
el\, J. L. (2004). On the identification of the pointwise Hölder exponent
of the generalized multifractional Brownian motion. Stochastic Processes
and their Applications\, 111(1)\, 119–56.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
14/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Nishika Ranathunga (La Trobe University)
DTSTART:20201209T010000Z
DTEND:20201209T020000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/15
DESCRIPTION:Title: Confidence Intervals in General Regression Models
that Utilize Uncertain Prior Information\nby Nishika Ranathunga (La Tr
obe University) as part of La Trobe University Statistics and Stochastic z
oom seminar\n\n\nAbstract\nWe consider a general regression model\, withou
t a scale parameter. We construct a confidence interval for a scalar param
eter of interest that utilizes the uncertain prior information that a dist
inct scalar parameter takes the specified value. This confidence interval
has good coverage properties. It also has scaled expected length\, where t
he scaling is with respect to the usual confidence interval\, that is (a)
substantially less than 1 when the prior information is correct\, (b) has
a maximum value that is not too large and (c) is close to 1 when the data
and prior information are highly discordant.\n\nFurthermore\, in Kabaila a
nd Ranathunga (2020)\, we solve the problem of numerically evaluating the
expected value of a smooth bounded function of a chi-distributed random va
riable\, divided by the square root of the number of degrees of freedom\,
using Mori's transformation followed by the trapezoidal rule\, which is ex
ponentially convergent for suitable integrands. This problem arises in sim
ultaneous inference\, selection and ranking of populations\, the evaluatio
n of multivariate t probabilities and the assessment of coverage and expec
ted volume properties of non-standard confidence regions.\n\nWe apply this
solution in the R package ciuupi2 that computes the Kabaila and Giri (200
9) confidence interval\, which utilizes the uncertain prior information in
a linear regression model with unknown error variance. Previous computati
ons of this interval used MATLAB programs that were time-consuming to run.
By writing these programs in R\, the computation time is greatly reduced
and they become freely available. We also assess a new definition of scale
d expected length.\n\nFinally\, we compare the computations of the log-lik
elihood function for generalized linear mixed models using (a) adaptive Ga
uss-Hermite quadrature and (b) importance sampling\, where both methods sh
are the same initial step (Kabaila and Ranathunga\, 2019).\n\nReferences:\
n\n Kabaila\, P.\, & Giri\, K. (2009). Confidence intervals in regressi
on utilizing prior information. Journal of Statistical Planning and Infere
nce\, 139\, 3419-3429.\n Kabaila P. and Ranathunga N. (2019) On Adaptiv
e Gauss-Hermite Quadrature for Estimation in GLMM’s. In: Nguyen H. (eds)
Statistics and Data Science. RSSDS 2019. Communications in Computer and I
nformation Science\, vol 1150. Springer\, Singapore.\n Kabaila\, P.\, &
Ranathunga\, N. (2020). Computation of the expected value of a chi-distri
buted random variable. Computational Statistics.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
15/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Jason Grealey (La Trobe University\, Baker Institute)
DTSTART:20210312T010000Z
DTEND:20210312T020000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/16
DESCRIPTION:Title: Quantifying computational carbon footprints and de
ep learning in genomic prediction\nby Jason Grealey (La Trobe Universi
ty\, Baker Institute) as part of La Trobe University Statistics and Stocha
stic zoom seminar\n\n\nAbstract\nThis presentation details the three main
projects undertaken within my PhD. The first involves investigating the ca
rbon footprint of computation. As climate change is an extremely pressing
global issue\, researchers must be prudent with energy usage\, this includ
es computational research. In this first project we developed a freely ava
ilable and simple to use carbon footprint estimator of computational tools
called Green Algorithms\, it provides interpretable metrics to understand
any given carbon footprint. The next section I will talk about involves t
he estimation of the carbon footprints of various bioinformatic analyses u
sing published benchmarks. These carbon footprints are largely unknown and
underappreciated within the research community\, we also provide a list o
f realistic and practical recommendations that computational researchers c
an utilise in order to minimise their carbon footprint. The last section i
s a simulation study aiming to understand what types of genetic architectu
res and study designs are needed to utilise neural networks in place of tr
aditional linear polygenic scoring methods in genomic prediction.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
16/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Nicholas Sing (La Trobe University\, Baker Institute)
DTSTART:20210401T010000Z
DTEND:20210401T020000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/17
DESCRIPTION:Title: Mining Lipidomics for Biological Insight\nby N
icholas Sing (La Trobe University\, Baker Institute) as part of La Trobe U
niversity Statistics and Stochastic zoom seminar\n\n\nAbstract\nLipidomics
is the study of all lipids that make up cells and organisms. Abnormal lip
id metabolism is associated with many cardiovascular risk factors. The Ba
ker Heart and Diabetes Institute has generated lipidomic datasets for seve
ral population studies. These datasets can contain hundreds of lipid speci
es and sample numbers ranging from hundreds to several thousands. During l
ipidomic analysis unwanted variation can arise due to variation from techn
ical sources\, which unwanted variation removal algorithms aim to minimise
. This project aims to develop multivariate methodologies for dealing with
unwanted variation in lipidomic datasets and modelling the metabolic asso
ciations between groups of lipid species and participant characteristics.
We intend to use eigenlipids to explore the existence\, onset or progressi
on of metabolic disease. We have demonstrated that eigenlipids can outperf
orm many individual lipid species in predicting cardiovascular risk factor
s. To identify technical sources of unwanted variation in the plasma lipid
ome during laboratory processing we recently performed a laboratory experi
ment\, which will support the utilisation of unwanted variation removal al
gorithms for removing variation from laboratory processing in pre-existing
datasets.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
17/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Mitra Jazayeri (La Trobe University)
DTSTART:20210506T020000Z
DTEND:20210506T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/18
DESCRIPTION:Title: Factors affecting first year psychology students
’ statistics learning\nby Mitra Jazayeri (La Trobe University) as pa
rt of La Trobe University Statistics and Stochastic zoom seminar\n\n\nAbst
ract\nTeaching statistics to different disciplines is increasingly challen
ging. This is due to several factors including the wide range of students
’ academic backgrounds\, availability of data and public perception of i
ts importance. In addition\, advancements in technology and recent technol
ogical innovations in teaching also present challenges due to the large ga
p between learning theory and teaching practices. \nFurthermore\, experie
ncing anxiety when studying statistics\, as a prerequisite subject\, has a
lways been commonplace for students around the world. Statistics anxiety c
an appear as a complex array of emotional reactions from only a minor disc
omfort to severe forms of apprehension\, fear\, nervousness\, panic and wo
rry. Considering that statistics is often required as a core subject in a
wide range of university degrees\, research into assisting in overcoming
these challenges is essential. \nThis research aims to explore interventi
on methods to minimize students’ apprehension in their learning process.
This is presented in three parts: 1) an examination of the effect of ble
nded delivery of an introductory statistics subject\, 2) a systematic revi
ew investigating interventions utilized to reduce students’ statistics a
nxiety\, 3a) the introduction of a survey tool for evaluation of student
attitudes\, confidence\, anxiety\, and beliefs about the usefulness of lea
rning statistics in their degree and an assessment of its’ reliability
and validity\, 3b) design\, implementation and analysis of a web-based min
dfulness intervention delivered to a sample of 530 students studying stati
stics for psychology during COVID-19 era. This project will help inform
educators for the better delivery of statistics to students with diverse a
cademic backgrounds.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
18/
END:VEVENT
BEGIN:VEVENT
SUMMARY:A/Prof Minh-Ngoc Tran (University of Sydney)
DTSTART:20210617T020000Z
DTEND:20210617T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/19
DESCRIPTION:Title: Variational Bayes on Manifolds\nby A/Prof Minh
-Ngoc Tran (University of Sydney) as part of La Trobe University Statistic
s and Stochastic zoom seminar\n\n\nAbstract\nVariational Bayes (VB) has be
come a widely-used tool for Bayesian inference in statistics and machine l
earning. Nonetheless\, the development of the existing VB algorithms is so
far generally restricted to the case where the variational parameter spac
e is Euclidean\, which hinders the potential broad application of VB metho
ds. This paper extends the scope of VB to the case where the variational p
arameter space is a Riemannian manifold. We develop an efficient manifold-
based VB algorithm that exploits both the geometric structure of the const
raint parameter space and the information geometry of the manifold of VB a
pproximating probability distributions. Our algorithm is provably converge
nt and achieves a decent convergence rate. We develop in particular severa
l manifold VB algorithms including Manifold Gaussian VB and Stiefel Neural
Network VB\, and demonstrate through numerical experiments that the propo
sed algorithms are stable\, less sensitive to initialization and compares
favourably to existing VB methods. This is a joint work with Dang Nguyen a
nd Duy Nguyen.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
19/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Masters Students (La Trobe University)
DTSTART:20210624T020000Z
DTEND:20210624T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/20
DESCRIPTION:Title: Masters Students Talks\nby Masters Students (L
a Trobe University) as part of La Trobe University Statistics and Stochast
ic zoom seminar\n\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
20/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr Matias Quiroz (University of Technology Sydney)
DTSTART:20210819T020000Z
DTEND:20210819T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/21
DESCRIPTION:Title: Spectral Subsampling MCMC for Stationary Multivari
ate Time Series\nby Dr Matias Quiroz (University of Technology Sydney)
as part of La Trobe University Statistics and Stochastic zoom seminar\n\n
\nAbstract\nSpectral subsampling MCMC was recently proposed to speed up Ma
rkov chain Monte Carlo (MCMC) for long stationary univariate time series b
y subsampling periodogram observations in the frequency domain. This talk
presents an extension of the approach to stationary multivariate time seri
es. We also propose a multivariate generalisation of the autoregressive te
mpered fractionally differentiated moving average model (ARTFIMA). The new
model is shown to provide a better fit compared to multivariate autoregre
ssive moving average models for three real world examples. We demonstrate
that spectral subsampling may provide up to two orders of magnitude faster
estimation\, while retaining MCMC sampling efficiency and accuracy\, comp
ared to spectral methods using the full dataset.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
21/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof Christopher Drovandi (Queensland University of Technology)
DTSTART:20210909T020000Z
DTEND:20210909T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/22
DESCRIPTION:Title: Statistical Inference for Implicit Models using Ba
yesian Synthetic Likelihood\nby Prof Christopher Drovandi (Queensland
University of Technology) as part of La Trobe University Statistics and St
ochastic zoom seminar\n\n\nAbstract\nImplicit models are defined as those
that can be simulated but the associated likelihood function is intractabl
e. Such models are prevalent in many fields such as biology\, ecology\, c
osmology and epidemiology. Given the unavailability of the likelihood fun
ction\, statistical inference for implicit models is challenging as we mus
t rely only on the ability to generate mock datasets from the model of int
erest\, and compare it with the observed data in some way. This talk will
explain a useful method called Bayesian synthetic likelihood for conducti
ng such statistical inference. I will discuss how BSL can be extended to
reduce the number of model simulations required and to make it more robust
to model misspecification. I will also describe some theoretical propert
ies of the method.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
22/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Illia Donhauzer (La Trobe University)
DTSTART:20210916T020000Z
DTEND:20210916T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/23
DESCRIPTION:Title: On the asymptotic behavior of functionals of rando
m fields\nby Illia Donhauzer (La Trobe University) as part of La Trobe
University Statistics and Stochastic zoom seminar\n\n\nAbstract\nThe talk
is about the asymptotic behavior of functionals of random fields with pos
sible long-range dependence. New properties of generalized Hermite-type pr
ocesses\, the Strong Law of Large Numbers (SLLN) for random fields\, and t
he asymptotic behavior of running maxima of random double arrays will be d
iscussed.\n\nNew properties of generalized Hermite-type processes that ari
se in NLT for integral functionals of long-range dependent random\n\nfield
s will be demonstrated. Contrary to the classical one-dimensional case\, i
t will be shown that for any choice of a multidimensional observation wind
ow the generalized Hermite-type process has non-stationary increments.\n\n
The SLLN for integral functionals of random fields with unboundedly increa
sing covariances will be presented. The SLLN is derived for the case of in
creasing domains. Conditions on covariance functions such that the SLLN ho
lds will be provided. The considered scenarios include non-stationary rand
om fields. The discussion about applications to weak and long-range depend
ent random fields and numerical examples will be shown.\n\nResults on the
asymptotic behavior of running maxima functionals of random double arrays
of phi-subgaussian random variables will be demonstrated. The main results
are specified for various important particular scenarios and classes of p
hi-subgaussian random variables.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
23/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr. Miryana Grigorova (University of Leeds)
DTSTART:20210930T090000Z
DTEND:20210930T100000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/24
DESCRIPTION:Title: Superhedging of options in a non-linear incomplete
financial market model.\nby Dr. Miryana Grigorova (University of Leed
s) as part of La Trobe University Statistics and Stochastic zoom seminar\n
\n\nAbstract\nWe will study the superhedging price (and superhedging strat
egies) of European and American options in a non-linear incomplete market
model with default\, with a particular focus on the American options case
which is more involved. We will provide a dual representation of the sell
er’s (superhedging) price for the American option in terms of a mixed st
ochastic control/stopping problem with non-linear expectations/ evaluation
s\, and in terms of non-linear Reflected BSDEs with constraints. If time p
ermits\, we will also present a duality result for the buyer’s price in
terms of a stochastic game of control and stopping with non-linear expecta
tions/ evaluations.\n\nZoom meeting link:\n\nhttps://unimelb.zoom.us/j/869
51431269?pwd=S1FPSFBHLzd5QkpGYlJIYS9wUGtLUT09\n\n(if the link doesn't work
when you click it -- please copy & paste it into the address bar in your
browser).\n\nPassword: 422668 (just in case)\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
24/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr. Francis K.C. Hui (Australian National University)
DTSTART:20211021T010000Z
DTEND:20211021T020000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/25
DESCRIPTION:Title: Spatio-temporal joint species distribution modelin
g – A community-level basis function approach\nby Dr. Francis K.C. H
ui (Australian National University) as part of La Trobe University Statist
ics and Stochastic zoom seminar\n\n\nAbstract\nThe last decade in ecology
has seen the development and rising popularity of joint species distributi
on modeling\napproaches for studying species assemblages\, with by far the
most common approach being based around\ngeneralized linear latent variab
le models (LVMs). However\, while methodological and computational advance
s\ncontinue to be made with LVMs\, their application to spatio-temporal mu
ltivariate abundance data i.e.\, observations\nof multiple species recorde
d across space and/or time\, remains computationally challenging and not n
ecessarily\nscalable when it comes to fitting and inference.\n\nIn this ta
lk\, we propose an alternative approach to spatio-temporal joint species d
istribution modeling which breaks\naway from the LVM framework. Inspired b
y the concept of fixed rank kriging\, we employ a set of fixed\, community
level\nspatial and/or temporal basis functions\, with corresponding specie
s-specific random slopes to account for\nspatio-temporal correlations both
within and between species. The resulting community-level basis function
model\n(CBFM) can be used for the same array of purposes as LVMs\, but is
designed to be computationally much more\nefficient given they can be set
up and thus fitted using the same machinery as for generalized additive mo
dels.\nSimulations and an application to a demersals fish dataset collecte
d off the Northeast US continental shelf illustrate\nthe potential of CBFM
s for scalable spatio-temporal joint species distribution modeling.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
25/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Masters Students (La Trobe University)
DTSTART:20211104T010000Z
DTEND:20211104T020000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/26
DESCRIPTION:Title: Masters Students Talks\nby Masters Students (L
a Trobe University) as part of La Trobe University Statistics and Stochast
ic zoom seminar\n\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
26/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Mitra Jazayeri (La Trobe University)
DTSTART:20211216T040000Z
DTEND:20211216T050000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/27
DESCRIPTION:Title: Validity analysis of a modified extended Technolog
y Acceptance Model for first year Psychology students\nby Mitra Jazaye
ri (La Trobe University) as part of La Trobe University Statistics and Sto
chastic zoom seminar\n\n\nAbstract\nThis talk continues on from my confirm
ation where an outline of my research and the results for the first two ph
ases of my project were given. In this talk I predominantly present the st
eps taken to evaluate the survey tool\, which is a modified and extended T
echnology Acceptance Model (TAM). This measurement scale determines the pe
rception of psychology students about the ease of use and usefulness of st
atistical concepts and their application in psychology using the statistic
al software\, SPSS.\n\nThe proposed model was tested for its reliability a
nd structural validity using data collected from a survey of first year ps
ychology students studying statistics during the global pandemic in 2020.
To explore the structure of the constructs of students’ attitude\, confi
dence and perception\, an Exploratory Factor Analysis (EFA) was conducted
on the responses data set. Five latent variables were identified. Utilizin
g maximum likelihood estimates in Confirmatory Factor Analysis (CFA)\, and
Analysis of Moment Structures (AMOS)\, the results supported the proposed
EFA model. In addition\, results of the CFA indicated that the best fitte
d model had correlations among four of the five constructs. Internal consi
stency estimates utilizing alpha coefficients\, ranged from 0.81 to 0.88 w
ith only one exception of 0.682. The findings provide a valid and reliable
assessment of students’ attitudes towards statistics for predicting aca
demic performance. Consequently\, this may help as a guide for effective d
ecision-making in the design and development of the statistics subjects fo
r students with a non-mathematical background.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
27/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Yeshna Bhowon (La Trobe Universty)
DTSTART:20220223T223000Z
DTEND:20220223T233000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/28
DESCRIPTION:Title: Applications Of Data Science Methods Within A Comm
unity-Based Physical Activity Program For Young People With Disability
\nby Yeshna Bhowon (La Trobe Universty) as part of La Trobe University Sta
tistics and Stochastic zoom seminar\n\n\nAbstract\nFitSkills is a communit
y-based program that connects university student mentors to young people l
iving with disability through exercise programs at their local community g
yms. Access to exercise facilities is a commonly documented perceived barr
ier to participation in physical activity for people living with disabilit
y\, but the problem has not been quantified. We conducted a geospatial ana
lysis using a population cohort in an aim to quantify this perceived barri
er. The second part of my research used data collected during the FitSkill
s trial to determine if completing FitSkills fostered positive attitudes t
owards disability among the student mentors.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
28/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Nicholas Sing (La Trobe University and Baker Heart and Diabetes In
stitute)
DTSTART:20220407T020000Z
DTEND:20220407T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/29
DESCRIPTION:Title: Mining Lipidomics for Biological Insight\nby N
icholas Sing (La Trobe University and Baker Heart and Diabetes Institute)
as part of La Trobe University Statistics and Stochastic zoom seminar\n\n\
nAbstract\nLipidomics is the study of all lipids that make up cells and or
ganisms. Abnormal lipid metabolism is associated with many cardiovascular
risk factors. The Baker Heart and Diabetes Institute has generated lipidom
ic datasets for several population studies. During lipidomic analysis unwa
nted variation can arise due to variation in laboratory processing and han
dling\, which unwanted variation removal algorithms aim to minimise. This
project aims to develop multivariate methodologies for dealing with unwant
ed variation in lipidomic datasets and modelling the metabolic association
s between groups of lipid species and participant characteristics. In this
project we are using lipid set enrichment analysis and eigenlipids to exp
lore lipid biology associated with cardiovascular disease. To identify tec
hnical sources of unwanted variation in the plasma lipidome during laborat
ory processing we performed a laboratory experiment and utilised the Remov
e Unwanted Variation-III (RUV-III) algorithm to remove these sources of un
wanted variation from the lipidomic dataset we acquired. We intend to use
this as a basis to identify negative control lipids to remove similar sour
ces of unwanted variation in population lipidomic datasets using RUV-III.\
n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
29/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr Alan Huang (University of Queensland)
DTSTART:20220428T020000Z
DTEND:20220428T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/30
DESCRIPTION:Title: On arbitrarily underdispersed discrete distributio
ns\nby Dr Alan Huang (University of Queensland) as part of La Trobe Un
iversity Statistics and Stochastic zoom seminar\n\n\nAbstract\nWe review a
range of generalized count distributions\, investigating which (if any) c
an be arbitrarily underdispersed\, i.e.\, its variance can be arbitrarily
small compared to its mean. A philosophical implication is that models fai
ling this criterion perhaps should not be considered a “statistical mode
l” according to the extendibility criterion of McCullagh (2002). Four pr
actical implications will be discussed. We suggest that all generalization
s of the Poisson distribution be tested against this property.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
30/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Ayesha Perera (La Trobe Univerity)
DTSTART:20220616T020000Z
DTEND:20220616T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/31
DESCRIPTION:Title: Performance of Model Averaged Tail Area Confidence
Interval\nby Ayesha Perera (La Trobe Univerity) as part of La Trobe U
niversity Statistics and Stochastic zoom seminar\n\n\nAbstract\nEvery mode
l has an uncertainty in the variables that it should include. Model averag
ing is considered as a promising method that could be used to perform infe
rence in the presence of model uncertainty. The performance of this method
heavily depends on the data-based model weights used. Traditionally\, thi
s weight is chosen to be proportional to the exponential of minus the Gene
ralized Information Criterion divided by two. We observe that the model-ba
sed confidence interval performs better\, in terms of coverage and expecte
d length\, in the case of two nested linear regression models when this di
vision by two is replaced by multiplied by a positive tuning constant. In
the second part of the talk\, we extend the analysis of the performance of
Model Averaged Tail Area confidence interval by Kabaila\, Welsh and Abeys
ekara\, Scandinavian Journal of Statistics\, 2016\, to the case of three o
r more nested linear regression models. We also assess the influence of th
e weight function on the performance of this confidence interval for three
nested linear regression models applied to the ‘Cholesterol’ data set
.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
31/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Masters students talks (La Trobe Univerity)
DTSTART:20220623T020000Z
DTEND:20220623T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/32
DESCRIPTION:Title: 2 masters theses presentation\nby Masters stud
ents talks (La Trobe Univerity) as part of La Trobe University Statistics
and Stochastic zoom seminar\n\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
32/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Faraz Fattah Hesari (La Trobe Univerity)
DTSTART:20220630T020000Z
DTEND:20220630T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/33
DESCRIPTION:Title: Statistical modelling of property prices\, applyin
g genetic algorithm and isolation forests\nby Faraz Fattah Hesari (La
Trobe Univerity) as part of La Trobe University Statistics and Stochastic
zoom seminar\n\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
33/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr Boris Buchmann (Australian National University)
DTSTART:20220804T070000Z
DTEND:20220804T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/34
DESCRIPTION:Title: Weak Subordination of Multivariate Levy Processes<
/a>\nby Dr Boris Buchmann (Australian National University) as part of La T
robe University Statistics and Stochastic zoom seminar\n\n\nAbstract\nYou
are welcome to attend the following Statistics and Stochastic colloquium (
part of the Colloquium Series of the Department of Mathematics and Statist
ics) at La Trobe Universit\, which is jointly organized with the Probabili
ty Victoria Seminar.\n\nPVSeminar #36\, Thursday 04 August / 17:00 AEST
\n\nBoris Buchmann (Australian National University\, Australia): Weak Subo
rdination of Multivariate Levy Processes \n\nAbstract: Subordination is th
e operation which evaluates a Levy process at a subordinator\, giving rise
to a pathwise construction of a "time-changed" process. Originating with
Bochner in the context of probability semigroups\, subordination was appli
ed by Madan and Seneta to create the variance gamma process\, which is pro
minently used in financial modelling. However\, unless the subordinate has
independent components or the subordinator has indistinguishable componen
ts\, subordination may not produce a Levy process. \n\nWe introduce a new
operation known as weak subordination that always produces a Levy process
by assigning the distribution of the subordinate conditional on the value
of the subordinator\, and matches traditional subordination in law in the
cases above. Weak subordination is applied to extend the class of varianc
e generalised gamma convolutions and to construct the weak variance-alpha-
gamma process. The latter process exhibits a wider range of dependence tha
n using traditional subordination. \n\nJoint work with Kevin W Lu (UW)\,
Dilip B Madan (UM)\, Marcus Michaelsen (UHH)\, Adam Nie (NTU)\, Alex Szima
yer (UHH). \n\nZoom meeting link: https://unimelb.zoom.us/j/83757047993?p
wd=a04zNitYZTRHdTZYdERkMmJYdDRWZz09\n \n(if the
link doesn't work when you click it -- please copy & paste it into the add
ress bar in your browser).\n\nPassword: 916563 (just in case)\n\nA PDF
file with the talk slides might become available for downloading from our
seminar Webpage at https://probvic.wordpress.com/pvseminar/ prior to the
talk (the above Zoom is being posted there).\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
34/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof Yimin Xiao (Michigan State University\, USA)
DTSTART:20220818T000000Z
DTEND:20220818T010000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/35
DESCRIPTION:Title: Sample Path and Extreme Value Properties of Multiv
ariate Gaussian Random Fields\nby Prof Yimin Xiao (Michigan State Univ
ersity\, USA) as part of La Trobe University Statistics and Stochastic zoo
m seminar\n\n\nAbstract\nIn this talk\, we present some recent results on
sample path and extreme value properties of a large class of multivariate
Gaussian random fields including multivariate Gaussian fields\, operator f
ractional Brownian motion\, vector-valued operator-scaling random fields\,
and matrix-valued Gaussian random fields. These results illustrate explic
itly the effects of the dependence structures among the coordinate process
es on the sample path and extreme value properties of multivariate Gaussia
n random fields.\n\nZoom meeting link: https://unimelb.zoom.us/j/864602693
83?pwd=aDNWbk4yWDdzclhUOWZ6ZElFQnlrQT09 \n \n(if
the above link doesn't work when you click it -- please copy & paste it i
nto the address bar in your browser).\n\nPassword: 457925 (just in case)\n
\nA PDF file with the talk slides might become available for downloading f
rom our seminar Webpage at https://probvic.wordpress.com/pvseminar/ prior
to the talk (the above Zoom link has already been posted there).\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
35/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof Giovanni Peccati (University of Luxembourg)
DTSTART:20220908T070000Z
DTEND:20220908T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/36
DESCRIPTION:Title: Some variations on a theme by P. de Jong\nby P
rof Giovanni Peccati (University of Luxembourg) as part of La Trobe Univer
sity Statistics and Stochastic zoom seminar\n\n\nAbstract\nJoint La Trobe
statistics and stochastics and PV seminar\n\nAbstract: In a remarkable pap
er from 1990\, the Dutch mathematician P. de Jong proved a striking Centra
l Limit Theorem yielding that\, for a sequence of normalized and degenerat
e U-statistics verifying a Lindeberg-type condition\, convergence to Gauss
ian is equivalent to the convergence of their fourth cumulants to zero. Su
ch a result is the ancestor of the collection of “fourth-moment theorems
” for non-linear functionals of random fields\, that have recently playe
d a prominent role in several questions of mathematical physics and stocha
stic geometry. In my talk\, I will first present some quantitative multidi
mensional extensions of de Jong’s result\, obtained by using Stein’s m
ethod of exchangeable pairs. I will then discuss some recent functional ve
rsions of de Jong’s findings\, both in the symmetric and non-symmetric c
ases. The results in the symmetric case yield some novel universality resu
lts for U-processes\, generalizing a classic invariance principle by Mille
r and Sen (1972)\, and allowing one to establish a complete taxonomy of fu
nctional CLTs associated with counting statistics of random geometric grap
hs. My presentation is mainly based on the following references:\n\nCh. D
öbler and G. Peccati: Quantitative de Jong Theorems in any dimension. EJP
\, 2016.\n\nCh. Döbler\, M. Kasprzak and G. Peccati: Weak convergence of
U-processes with size-dependent kernels. Ann. App. Prob.\, 2022\n\nCh. Dö
bler\, M. Kasprzak and G. Peccati. The multivariate functional de Jong CLT
. Probab. Th. Rel. Fields\, 2022+\n\nZoom meeting link: https://unimelb.zo
om.us/j/82317899187?pwd=TThhQmZrcGtxSGpQL2wzTHJjZlZjQT09\n
\n(if the above link doesn't work when you click it -- please c
opy & paste it into the address bar in your browser).\n\nPassword: 633070
(just in case)\n\nA PDF file with the talk slides might become available f
or downloading from our seminar Webpage at https://probvic.wordpress.com/p
vseminar/ prior to the talk (the above Zoom link will also be posted there
shortly).\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
36/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Honours and Masters students (La Trobe University)
DTSTART:20221027T010000Z
DTEND:20221027T023000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/37
DESCRIPTION:Title: Honours and Masters theses presentations\nby H
onours and Masters students (La Trobe University) as part of La Trobe Univ
ersity Statistics and Stochastic zoom seminar\n\n\nAbstract\nThe schedule
is the following (20min for Thesis B talks and 10min for Thesis A talk):\n
\n\n12.05pm Adam Bilchouris. Investigating Statistical Properties of Funct
ionals of Strongly Dependent Spatial Data.\n\n12.15pm Lennon Zachary Logan
. Mathematics of Kirigami. A Study of Euclidean Nets and Hyperbolic Crysta
ls.\n\n12.35pm Dmytro Ostapenko. Statistical Modelling of ANZ Property Dat
a.\n\n12.55pm Juliet Nwabuzor. Performance of Preliminary Model Selection
Using Bayesian Information Criterion (Bic).\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
37/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Mitra Jazayeri (La Trobe university)
DTSTART:20221214T233000Z
DTEND:20221215T003000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/38
DESCRIPTION:Title: Teaching statistics in a new era- factors affectin
g students’ learning and aligning technology with how we teach\nby M
itra Jazayeri (La Trobe university) as part of La Trobe University Statist
ics and Stochastic zoom seminar\n\n\nAbstract\nThis thesis investigates th
e factors that affect psychology students’ ability to learn statistics i
n a new era of technological advancements and the effects post-Covid on ed
ucation. Historically\, teaching statistics to psychology students has bee
n one of the most challenging tasks for statistics educators worldwide. Pr
evious research has been conducted around the theme of social science stud
ents’ statistics anxiety and the varied survey designs developed to meas
ure it. However\, little has been done to reduce statistics anxiety\, with
the aim of increasing the performance of students with a non-mathematical
background\, particularly in this technological age.\n\n \nThe aim of th
is research is to i) conduct multiple regression and sub-group analyses us
ing the R software package to investigate whether the blended delivery of
a 12-week statistics subject to first-year psychology students had any eff
ect on performance compared to face-to-face teaching only\; ii) design a m
indfulness intervention\, together with a step-by-step methodological appr
oach for teaching statistics to first-year psychology students\; iii) dev
elop a survey based on the technology acceptance model to measure students
’ anxiety which included testing the validity and reliability of the ado
pted survey tool. To do so\, the structural properties of the survey were
investigated. For this stage of the research\, jamovi and the IBM SPSS AMO
S software package were utilized to obtain Cronbach’s alpha and the expl
oratory and confirmatory factor analysis output. This thesis finds that th
e web-based mindfulness intervention had a significant positive effect on
students’ statistics anxiety and therefore performance. Moreover\, five
constructs were identified which affect students’ statistics anxiety and
therefore their performance\, namely attitude\, confidence\, student’s
awareness of their mental state\, independent learner belief\, and depende
nt learner belief. The findings of this research may assist and inspire st
atistics educators internationally in their approach to the design and dev
elopment of their teaching material to non-mathematical students for whom
statistics is a core subject in their study.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
38/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof. Gennady Samorodnitsky (Cornell University)
DTSTART:20230322T230000Z
DTEND:20230323T000000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/39
DESCRIPTION:Title: Clustering of large deviations in moving average p
rocesses: short and long memory regimes\nby Prof. Gennady Samorodnitsk
y (Cornell University) as part of La Trobe University Statistics and Stoch
astic zoom seminar\n\n\nAbstract\nYou are welcome to attend the following
Statistics and Stochastic colloquium (part of the Colloquium Series of the
Department of Mathematics and Statistics) at La Trobe University organise
d together with the PVSeminar (please note the UNUSUAL time).\n\n\n+++++++
+++++++++++\n\nPVSeminar #44\, Thursday 23 March / 10:00 AEDT \n\nGennady
Samorodnitsky (Cornell University\, United States of America): Clustering
of large deviations in moving average processes: short and long memory re
gimes\n \nAbstract: We describe the cluster of large deviations events tha
t arise when one such large deviations event occurs. We work in the framew
ork of an infinite moving average process with a noise that has finite exp
onential moments. The cluster turns out to have different shapes in the ca
ses when the moving average process has short memory and long memory.\n\nJ
oint work with Arijit Chakrabarty.\n\nZoom meeting link: \n\nhttps://unime
lb.zoom.us/j/88379660402?pwd=bzh6WUM3UFR5dUhnVjFQdWhUOXlCZz09\n\n\n(if the
above link doesn't work when you click it -- please copy & paste it into
the address bar in your browser).\n\nPassword: 005582 (just in case)\n\nA
PDF file with the talk slides might become available for downloading from
our seminar Webpage at https://probvic.wordpress.com/pvseminar/ prior to t
he talk (the above Zoom link is already there).\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
39/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof. V.S. Matveev (Die Friedrich-Schiller-Universität Jena\, the
Federal Republic of Germany // La Trobe University\, the Commonwealth of
Australia)
DTSTART:20230511T070000Z
DTEND:20230511T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/40
DESCRIPTION:Title: Geodesic random walks\, diffusion processes\, and
Brownian motion on Finsler manifolds\nby Prof. V.S. Matveev (Die Fried
rich-Schiller-Universität Jena\, the Federal Republic of Germany // La Tr
obe University\, the Commonwealth of Australia) as part of La Trobe Univer
sity Statistics and Stochastic zoom seminar\n\n\nAbstract\nWe show that ge
odesic random walks on a complete Finsler manifold of bounded geometry con
verge to a diffusion process which is\, up to a drift\, the Brownian motio
n corresponding to a Riemannian metric. In particular\, the generator of t
he limit process is a non-degenerate elliptic second-order partial differe
ntial operator for which we give a precise integral formula. If the geodes
ic random walk is geometric\, that is if the law of increments are constru
cted by the Finsler metric by a coordinate-invariant procedure\, the Riema
nnian metric is then determined by the Finsler metric. Special cases of su
ch functors F → g_F include the Binet-Legendre metric and different a
verage metrics and has many effective applications in Finsler geometry in
which in particular certain mathematicians from the La Trobe University ar
e involved. Also\, possible applications in natural and life sciences will
be discussed.\n\nMost results are joint with Tianyu Ma and Ilya Pavlyukev
ich.\n\nZoom meeting link: \n\nhttps://unimelb.zoom.us/j/86301899265?pwd=Z
mRqWVBDRzM1azlpRjVHWG5HaEZOUT09\n\n(if the above link doesn't work when yo
u click it -- please copy & paste it into the address bar in your browser)
.\n\nPassword: 277885 (just in case)\n\n\nA PDF file with the talk slides
(if any) might become available for downloading from the Webpage at https:
//probvic.wordpress.com/pvseminar/ prior to the talk.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
40/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof Ben Goldys (The University of Sydney\, the Commonwealth of Au
stralia)
DTSTART:20230525T070000Z
DTEND:20230525T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/41
DESCRIPTION:Title: Stochastic flows in infinite dimensions\nby Pr
of Ben Goldys (The University of Sydney\, the Commonwealth of Australia) a
s part of La Trobe University Statistics and Stochastic zoom seminar\n\n\n
Abstract\nStochastic flows associated with finite dimensional SDEs and ran
dom dynamical systems are an important tool for the study of large time be
haviour of solutions and for their pathwise analysis. The existence of (sm
ooth) stochastic flows in finite dimensions is well understood\, see for e
xample books by Kunita. The famous Skorokhod example shows that the situat
ion is completely different in infinite dimensions (hence for stochastic P
DEs). The existence of the flow for infinite dimensional stochastic system
s is known in some special cases but is little understood. We will present
new results on the existence of the flows in infinite dimensions. We will
also present applications to the existence theory of stochastic PDEs and
to the question of regularity of transition semigroups.\n\nZoom meeting li
nk: \n\nhttps://unimelb.zoom.us/j/81406950751?pwd=dWFUV0VpUmQyOHdtL0ZuUXcz
MTJ4Zz09\n\n(if the above link doesn't work when you click it -- please co
py & paste it into the address bar in your browser).\n\nPassword: 931016 (
just in case)\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
41/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Nicholas Sing (La Trobe University)
DTSTART:20230601T020000Z
DTEND:20230601T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/42
DESCRIPTION:Title: Mining Lipidomics for Biological Insight\nby N
icholas Sing (La Trobe University) as part of La Trobe University Statisti
cs and Stochastic zoom seminar\n\n\nAbstract\nLipidomics is the study of a
ll lipids that make up cells and organisms. Abnormal lipid metabolism is a
ssociated with many cardiovascular risk factors. Past lipidomic studies ha
ve lacked appropriate cohort sizes to explore bioinformatics methodologies
used for dealing with large cohorts. The Baker Heart and Diabetes Institu
te has uniquely generated lipidomic datasets for large population cohort s
tudies. During lipidomic analysis unwanted variation can arise due to vari
ation in laboratory processing and handling\, which unwanted variation rem
oval algorithms aim to minimise. To identify technical sources of unwanted
variation in the plasma lipidome during laboratory processing we performe
d a laboratory experiment and utilised the Removing Unwanted Variation-III
(RUV-III) algorithm to remove these sources of unwanted variation from (1
) the lipidomic dataset we acquired and (2) these same sources of unwanted
variation in a large population cohort study generated via the Baker Hear
t and Diabetes Institute’s Lipidomic’s platform. We then focus on mode
lling the metabolic associations between groups of lipid species and cardi
ovascular disease. We specifically\, used lipid set enrichment analysis an
d eigenlipids to explore lipid biology associated with cardiovascular dise
ase using the lipidomic datasets generated for two large population cohort
s via the Baker Heart and Diabetes Institute’s Lipidomics platform.\n\n\
n\nJoin from a PC\, Mac\, iOS or Android: https://latrobe.zoom.us/j/876085
35816\n\nOr iPhone one-tap (Australia Toll): +61280152088\,87608535816#\n
\nOr Telephone:\n Dial: +61 2 8015 2088\n Meeting ID: 876 0853 5816
\n International numbers available: https://latrobe.zoom.us/u/kNXgIF2zg
\n\nOr a H.323/SIP room system:\n Dial: 87608535816@zoom.aarnet.edu.au\
n or 87608535816@zmau.us\n or 103.122.166.55\n Meeting ID: 876085
35816\n\nOr Skype for Business (Lync):\n SIP:87608535816@lync.zoom.us\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
42/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Honours and Masters students (La Trobe University)
DTSTART:20230615T020000Z
DTEND:20230615T033000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/43
DESCRIPTION:Title: Honours and Masters theses presentations\nby H
onours and Masters students (La Trobe University) as part of La Trobe Univ
ersity Statistics and Stochastic zoom seminar\n\n\nAbstract\nThe schedule
is the following (20min for Thesis B talks and 10min for Thesis A talk):\n
\n12:00 Sanjanaa Roopa Sunil. Modelling the impact of the increase in huma
n lifespan on the emergence of infectious diseases.\n\n12:15 Adam Brown-Sa
rre. Metrical Theory of Continued Fractions.\n\n12:40 Adam Bilchouris. Met
hodologies for Exploring Out-of-Sample Prediction and Spatial Dependency f
or Complex Big Data.\n\n13:05 Nazmi Amir. Generalised Entropy Indices for
Analysis of Customers/Experts Opinion.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
43/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof. Antoine Ayache (Lille University\, France)
DTSTART:20230808T030000Z
DTEND:20230808T040000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/44
DESCRIPTION:Title: Harmonizable Fractional Stable Motion: simultaneou
s estimators for both parameters\nby Prof. Antoine Ayache (Lille Unive
rsity\, France) as part of La Trobe University Statistics and Stochastic z
oom seminar\n\n\nAbstract\nThere are two classical very different extensio
ns of the well-known Gaussian fractional Brownian motion to non-Gaussian f
rameworks of heavy-tailed stable distributions: \nthe harmonizable fractio
nal stable motion (HFSM) and the linear fractional stable motion (LFSM). A
s far as we know\, while several articles in the literature\, some of whic
h \nappeared a long time ago\, have proposed statistical estimators for th
e parameters of LFSM\, no estimator has yet been proposed in the framework
of HFSM. Among other \nthings\, what makes statistical estimation of para
meters of HFSM to be a difficult problem is that\, in contrast to LFSM\, H
FSM is not ergodic. The main goal of our talk is to \npropose a new strate
gy for dealing with this problem and obtaining solutions of it. The keysto
ne of our new strategy consists in the construction of new transforms of H
FSM \nwhich allow to obtain\, at any dyadic level\, a sequence of independ
ent stable random variables.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
44/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Maha Alghamdi (La Trobe University)
DTSTART:20230907T070000Z
DTEND:20230907T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/45
DESCRIPTION:Title: Properties of Functionals of Random Fields with Ap
plications to Spatial Data Analysis\nby Maha Alghamdi (La Trobe Univer
sity) as part of La Trobe University Statistics and Stochastic zoom semina
r\n\n\nAbstract\nIn this presentation\, I will overview the theory of stoc
hastic processes and random fields. I will discuss the historical developm
ent and key results of limit theorems for non-linear functionals of random
fields. Definitions of Hermite expansions\, semi-stable random fields and
slowly varying functions will be presented. Additionally\, I will highlig
ht a recent new result in the field of limit theorems for Gaussian spatial
processes\, specifically the Spectral Central Limit Theorem for Function
als of Isotropic and Stationary Gaussian Fields. The plan for the future r
esearch will be presented.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
45/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof Ivan Nourdin (University of Luxembourg)
DTSTART:20231116T073000Z
DTEND:20231116T083000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/46
DESCRIPTION:Title: Quantitative CLTs in Deep Neural Networks\nby
Prof Ivan Nourdin (University of Luxembourg) as part of La Trobe Universit
y Statistics and Stochastic zoom seminar\n\n\nAbstract\nAbstract: In this
talk\, we will study the distribution of a fully connected neural network
with random Gaussian weights and biases in which the hidden layer widths a
re proportional to a large constant n. More precisely\, we will explain ho
w to prove quantitative bounds on normal approximations valid at large but
finite n and any fixed network depth. This is based on a joint work with
S. Favaro\, B. Hanin\, D. Marinucci and G. Peccati.\n\nZoom meeting link:
\n\nhttps://unimelb.zoom.us/j/82073536928?pwd=b3dmOUFwdFNUZS9hMWxHZkFRV202
Zz09\n\n(if the above link doesn't work when you click it -- please copy &
paste it into the address bar in your browser).\n\nPassword: 094232 (jus
t in case)\n\nA PDF file with the talk slides (if any) might become availa
ble for downloading from our seminar web page at https://probvic.wordpress
.com/pvseminar/ prior to the talk (the above Zoom link has already been po
sted there).\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
46/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Masters students (La Trobe University)
DTSTART:20231116T010000Z
DTEND:20231116T020000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/47
DESCRIPTION:Title: Theses talks\nby Masters students (La Trobe Un
iversity) as part of La Trobe University Statistics and Stochastic zoom se
minar\n\n\nAbstract\nMasters theses presentations: \n\n12:00 Sanjanaa Roop
a Sunil. How might the change in human longevity have affected the emergen
ce and transmission of infectious diseases (Thesis B)\n\n12:25 Sal Sabila.
Modelling and Analysis of Property Prices Using Their Features (Thesis A)
\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
47/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Nemini Samarakoon (La Trobe University)
DTSTART:20231122T073000Z
DTEND:20231122T083000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/48
DESCRIPTION:Title: Construction\, Properties and Simulation of Haar-B
ased Multifractional Processes\nby Nemini Samarakoon (La Trobe Univers
ity) as part of La Trobe University Statistics and Stochastic zoom seminar
\n\nAbstract: TBA\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
48/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Adam Bilchouris (La Trobe University)
DTSTART:20240411T020000Z
DTEND:20240411T023500Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/49
DESCRIPTION:Title: Methods for Exploring Out-of-Sample Prediction and
Spatial Dependency for Complex Big Data\nby Adam Bilchouris (La Trobe
University) as part of La Trobe University Statistics and Stochastic zoom
seminar\n\n\nAbstract\nIn this presentation\, estimators of stochastic pr
ocesses and isotropic Gaussian random fields will be considered\, in parti
cular\, the covariance functions of these objects. The parameters of a Geg
enbauer process will be estimated through its spectral density\, and for i
sotropic Gaussian random fields\, the normality of the first Minkowski fun
ctional is tested. Several existing covariance function estimators will be
introduced\, and their properties will be discussed. New estimation techn
iques will be introduced for isotropic Gaussian random fields which were d
eveloped to handle denser and larger data sets (i.e. big data). Simulation
studies were conducted in order to determine the effectiveness of the abo
ve estimators under varying degrees of dependencies in the data.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
49/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Shahid Khan (La Trobe University)
DTSTART:20240411T023500Z
DTEND:20240411T031000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/50
DESCRIPTION:Title: Spherical Stochastic Models and Approximation Sche
mes\nby Shahid Khan (La Trobe University) as part of La Trobe Universi
ty Statistics and Stochastic zoom seminar\n\n\nAbstract\nIn numerous appli
cations it is important to model and analyse spatial data on a sphere. A c
ommon assumption is that spatial covariance functions exhibit isotropy. Wh
ile existing approaches in the literature focus on Euclidean spaces\, they
are not suitable for spherical domains. Utilizing spherical harmonic (SH)
representation\, we formulate a test for isotropy. Data is projected onto
spherical harmonic functions which form orthogonal basis functions on the
sphere. We then exploit the fact that if the process is isotropic\, the c
orrelation between the coefficients will be zero. This motivates a test ba
sed on the sample correlation matrix of SH coefficients\, using the larges
t eigenvalue as the test statistic. We test our method on simulated and Co
smic Microwave Background (CMB) radiation data by selecting random regions
and using CMB radiation data from them.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
50/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Jayamini Chamindira Liyanage (La Trobe University)
DTSTART:20240306T040000Z
DTEND:20240306T044000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/51
DESCRIPTION:Title: Multivariate Meta-Analysis Methods for High-Dimens
ional Data\nby Jayamini Chamindira Liyanage (La Trobe University) as p
art of La Trobe University Statistics and Stochastic zoom seminar\n\nAbstr
act: TBA\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
51/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr Selvaraaju Murugesan (SaaS startup Kovai.co)
DTSTART:20240423T020000Z
DTEND:20240423T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/52
DESCRIPTION:Title: Recent Trends in Data Science and AI\nby Dr Se
lvaraaju Murugesan (SaaS startup Kovai.co) as part of La Trobe University
Statistics and Stochastic zoom seminar\n\n\nAbstract\nArtificial Intellige
nce technology is now becoming ubiquitous. This technology landscape is sh
ifting rapidly given the rise of GenAI based apps such as ChatGPT\, Gemini
and so on. This talk will introduce students to recent trends in data sci
ence practice and Artificial Intelligence. The key takeaways would be\na.
Understand the importance of data and its business use case\nb. Know trend
s in AI\nc. Skillsets required to be a data scientist\nd. Building GenAI a
pps\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
52/
END:VEVENT
BEGIN:VEVENT
SUMMARY:A/Prof Simon Harris (University of Auckland)
DTSTART:20240502T070000Z
DTEND:20240502T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/53
DESCRIPTION:Title: Genealogies of samples from stochastic population
models\nby A/Prof Simon Harris (University of Auckland) as part of La
Trobe University Statistics and Stochastic zoom seminar\n\n\nAbstract\nYou
are welcome to attend the following Statistics and Stochastic colloquium
(part of the Colloquium Series of the Department of Mathematics and Statis
tics) at La Trobe University organised together with the PVSeminar (please
note the UNUSUAL time).\n\n++++++++++++++++++\n\nWhat does the family tre
e look like for a random sample of k individuals taken from some populatio
n? Surprisingly\, until relatively recently this fundamental question rema
ined an open problem even for sampling extant individuals in one of the si
mplest of stochastic population models. We will discuss some recent progr
ess in this area\, including the emergence of certain universal limiting g
enealogies when sampling individuals uniformly at random at large times fr
om large stochastically evolving populations\, such as near-critical Bieny
ame-Galton-Watson stochastic branching processes conditioned to survive. W
hilst these universal genealogical trees have the same tree topology as a
Kingman coalescent\, it turns out that their coalescent (or split) times a
re quite different due to stochastic population size effects\, although wi
th an explicit representation as some mixture of IID split times. We will
mention how these universal genealogies are intimately related to sampling
from the Brownian excursion and Aldous’ Continuum Random Tree. Some ong
oing research and open problems will also be discussed.\n\nThis talk is re
lated to various collaborations with S.Bocharov (Xi’an Jiaotong-Liverpoo
l)\, S.Johnston (KCL)\, J.C.Pardo (CIMAT)\, S.Palau (UNAM)\, and M.Roberts
(Bath). We also acknowledge the support of the New Zealand Royal Society
Te Apārangi Marsden fund.\n\nZoom meeting link: \n\nhttps://unimelb.zoom.
us/j/89920990824?pwd=aWhJNkgyMjFXU09JTnFZeFQ4d2J0QT09\n\n(if the above lin
k doesn't work when you click it -- please copy & paste it into the addres
s bar in your browser).\n\nPassword: 754118 (just in case)\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
53/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof D. Marinucci (University of Rome “Tor Vergata”)
DTSTART:20240523T060000Z
DTEND:20240523T070000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/54
DESCRIPTION:Title: Geometry and topology of spin random fields\nb
y Prof D. Marinucci (University of Rome “Tor Vergata”) as part of La T
robe University Statistics and Stochastic zoom seminar\n\n\nAbstract\nThis
event is being organized in collaboration with the PVSeminar.\n\nAbstract
: The investigation of the geometric properties of the excursion sets for
isotropic random fields has been the object of a considerable amount of re
search over the last decade. In this talk\, we discuss how to generalize s
ome of these results to the case of random fields which are not scalar val
ued\, but they are rather random sections of a non-trivial fiber bundle: t
hese are the so-called spin (spherical) random fields\, which play a key r
ole in Cosmology\, especially in connection with the analysis of Cosmic Mi
crowave Background (CMB) polarization. In particular\, we discuss how to c
haracterize their expected geometry and topology: i.e.\, we investigate th
e asymptotic behaviour\, under scaling assumptions\, of general classes of
functionals of (properly defined) excursion sets\, including Lipschitz-Ki
lling Curvatures and Betti Numbers.\n\nBased on joint works with Antonio L
erario\, Maurizia Rossi and Michele Stecconi.\n\nZoom meeting link: \n\nht
tps://unimelb.zoom.us/j/85717338005?pwd=ZzZLMmVCRjA5OU1ycS9iWWFPb1ptZz09\n
\n(if the above link doesn't work when you click it -- please copy & paste
it into the address bar in your browser).\n\nPassword: 122074 (just in ca
se)\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
54/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof Enzo Orsingher (L'Università degli Studi di Roma "La Sapienz
a")
DTSTART:20240509T070000Z
DTEND:20240509T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/55
DESCRIPTION:Title: Planar random motions at finite velocity\nby P
rof Enzo Orsingher (L'Università degli Studi di Roma "La Sapienza") as pa
rt of La Trobe University Statistics and Stochastic zoom seminar\n\n\nAbst
ract\nThis event is being organized in collaboration with the PVSeminar.\n
\nAbstract: In our talk\, we examine various models of motion in the plane
with a finite or an infinite number of directions\, The case of finite di
rections includes cyclic motions and clockwise or counterclockwise motions
when from a direction d_k one can pass either to direction d_(k+1) or d_(
k-1) with equal or different probabilities. Particular attention is paid t
o the planar motion with orthogonal directions. The switches of directions
are governed by a homogeneous or non-homogeneous Poisson process\, becaus
e in this case explicit results can be obtained. Planar motions with an in
finite number of directions are performed at velocity c with changes occur
ring at Poisson-paced times and each new direction is taken with a uniform
ly distributed angle. The minimal case (three directions) is also consider
ed and its connection with higher-order Bessel functions is presented.\n\n
Some of the results are published in: Fabrizio Cinque and Enzo Orsingher\,
Stochastic Dynamics of Generalized Planar Random Motions with Orthogonal
directions\, J. Theor. Probab. 2023\, 36(4)\, pp. 2229-2261\, and in the r
eferences of this paper\,\n\nZoom meeting link: \n\nhttps://unimelb.zoom.u
s/j/82613441430?pwd=d2hCQzJjUXVUelozMjlVcXo2K3hZQT09\n\n(if the above link
doesn't work when you click it -- please copy & paste it into the address
bar in your browser).\n\nPassword: 427391 (just in case)\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
55/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof Günter Last (Karlsruher Institut für Technologie)
DTSTART:20240613T070000Z
DTEND:20240613T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/56
DESCRIPTION:Title: Cluster density and uniqueness of the infinite clu
ster of the random connection\nby Prof Günter Last (Karlsruher Instit
ut für Technologie) as part of La Trobe University Statistics and Stochas
tic zoom seminar\n\n\nAbstract\nWe consider a random connection model (RC
M) on a general space driven by a Poisson process whose intensity measure
is scaled by a parameter $t\\ge 0$. An important special case is the stati
onary marked RCM (in Euclidean space)\, containing the Boolean model with
general compact grains and the so-called weighted RCM as special cases. We
say that the infinite clusters are deletion stable if the removal of a Po
isson point cannot split a cluster in two or more infinite clusters. We pr
ove that this stability together with a natural irreducibility assumption
implies uniqueness of the infinite cluster. We then show that the infinit
e clusters of the stationary marked RCM are deletion stable. It follows th
at an irreducible stationary marked RCM can have at most one infinite clus
ter which extends and unifies several results in the literature. An import
ant ingredient of our proofs are differentiability and convexity propertie
s of the cluster density which are of interest in their own right.\n\nThe
talk is based on recent joint work with Mikhail Chebunin: https://arxiv.or
g/abs/2403.17762. Some of the main ideas come from a seminal paper by Aize
nman\, Kesten and Newman (1987)\, treating discrete percolation models.\n\
njoint La Trobe Stochastics and Statistics and PV seminar\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
56/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Masters and Honours students
DTSTART:20240620T010000Z
DTEND:20240620T040000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/57
DESCRIPTION:Title: Masters and Honours theses talks\nby Masters a
nd Honours students as part of La Trobe University Statistics and Stochast
ic zoom seminar\n\n\nAbstract\n11:00 Lauren White. Continued Fractions Res
tricted to Prime Partial Quotients.\n11:15 Sigfrido Ciletti. Zero-One Laws
for Intuitionistic Logic.\n11:40 Sal Sabila. Modeling and Analysis of Pro
perty Prices Using Their Features.\n12:05 Tracey Anderson. Pseudorandom Ge
neration of Convex Lattice Polygons.\n12:30 Qian Ding. Ranking With Domina
nce Matrices: Dominance Quantification and Weighting Methods.\n12:55 Moham
med Mohsin Ghori. Hypergraphs and Lotka-Volterra Systems.\n13:20 Ryan Nahk
uri. Methods for Combining P-Values: A Comparison and a Tool for Researche
rs.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
57/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Maha Alghamdi (La Trobe University)
DTSTART:20240912T060000Z
DTEND:20240912T070000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/58
DESCRIPTION:Title: Multiscaling limit theorems for stochastic FPDE wi
th cyclic long-range dependence\nby Maha Alghamdi (La Trobe University
) as part of La Trobe University Statistics and Stochastic zoom seminar\n\
n\nAbstract\nThe talk will discuss solutions of stochastic partial differe
ntial equations with random initial conditions. First\, it overviews some
of the known results on scaled solutions of such equations and provides se
veral explicit examples. Then\, it presents multiscaling limit theorems fo
r renormalized solutions for the case of initial conditions subordinated t
o the random processes with cyclic long-range dependence. Two cases of sto
chastic partial differential equations are examined. The spectral and cova
riance representations for the corresponding limit random fields are obtai
ned. Additionally\, we will discussed why analogous results are not valid
for subordinated cases with Hermite ranks greater than 1. Numerical examp
les that illustrate the obtained theoretical results will presented.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
58/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Master students (La Trobe University)
DTSTART:20241114T010000Z
DTEND:20241114T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/59
DESCRIPTION:Title: Thesis talks\nby Master students (La Trobe Uni
versity) as part of La Trobe University Statistics and Stochastic zoom sem
inar\n\n\nAbstract\nLauren White\, Continued Fractions Restricted to Prime
Partial Quotients\nRyan Nahkuri\, Methods for Combining P-Values: A Compa
rison and a Tool for Researchers\nTracey Anderson\, Pseudorandom Generatio
n of Convex Lattice Polygons\nMohammed Mohsin Ghori\, Hypergraphs and Lotk
a-Volterra Systems\nQian Ding\, Ranking With Dominance Matrices: Dominance
Quantification and Weighting Methods\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
59/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Nemini Yadeesha Samarakoon (La Trobe University)
DTSTART:20250131T070000Z
DTEND:20250131T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/60
DESCRIPTION:Title: Construction\, Properties and Simulation of Haar-B
ased Multifractional Processes with Rmultifractional R package\nby Nem
ini Yadeesha Samarakoon (La Trobe University) as part of La Trobe Universi
ty Statistics and Stochastic zoom seminar\n\n\nAbstract\nMultifractional p
rocesses were introduced in the mid-1990s\, as the fractional Brownian mot
ion (fBm) could not model changes in the roughness of trajectories over ti
me. The constant Hurst parameter of fBm was replaced by a time-varying Hur
st function. The talk will introduce a new class of multifractional proces
ses\, the Gaussian Haar-based multifractional processes (GHBMP) which is b
ased on the Haar wavelet approach. GHBMP provides a theoretical model and
simulation tools for a wider class of processes and Hurst functions. The t
heoretical properties of the GHBMP will also be discussed. Additionally\,
Rmultifractional R package will be introduced. This R package includes fun
ctions for the simulation of Gaussian Haar-based multifractional processes
\, estimation of the Hurst function and analysis of multifractional proces
ses.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
60/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Shahid Khan (La Trobe University)
DTSTART:20250501T020000Z
DTEND:20250501T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/61
DESCRIPTION:Title: Exploring the Structure of Anisotropic Random Fiel
ds on the Sphere\nby Shahid Khan (La Trobe University) as part of La T
robe University Statistics and Stochastic zoom seminar\n\n\nAbstract\nThis
talk explores fundamental questions about the existence of isotropic and
anisotropic spherical fields and the behaviour of isotropic fields under c
omposite transformations. It identifies feasible and infeasible scenarios\
, highlighting differences from random fields in Euclidean spaces. Connect
ions to the spectral representations of spherical random fields are discus
sed. The results offer guidance for developing models and sampling strateg
ies for parameter estimation and hypothesis testing of spherical data.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
61/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Masters and Honors students (La Trobe University)
DTSTART:20250619T010000Z
DTEND:20250619T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/62
DESCRIPTION:Title: Masters and Honors thesis talks\nby Masters an
d Honors students (La Trobe University) as part of La Trobe University Sta
tistics and Stochastic zoom seminar\n\n\nAbstract\nPresentation Schedule:\
n\n11:00 Krishna Suppiah. Study of Gegenbauer Time Series and Their Statis
tics\n\n11:15 David Teakle. Generalised adaptive signed correlation index
\n\n11:40 Aarjav Khara. Correcting Predictions for Anomalies: Application
to Property Prices\n\n12:05 Helen Arnold. Prediction intervals in meta-ana
lysis\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
62/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Adam Bilchouris (La Trobe University)
DTSTART:20250814T070000Z
DTEND:20250814T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/63
DESCRIPTION:Title: Investigating Complex Spatial and Time Series Data
\nby Adam Bilchouris (La Trobe University) as part of La Trobe Univers
ity Statistics and Stochastic zoom seminar\n\n\nAbstract\nSeveral nonparam
etric estimators of autocovariance function can be found in the literature
with various statistical properties and potential drawbacks. Despite thei
r limitations\, these estimators remain widely used in many applications.\
nAn R package\, CovEsts\, will be introduced. The package provides several
nonparametric estimators of the autocovariance function with differing pr
operties\, through a unified interface and output. The package also includ
es estimator transforms\, estimator corrections\, and several metric funct
ions to compare obtained estimates.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
63/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof Adam Jakubowski (Uniwersytet Mikołaja Kopernika)
DTSTART:20250821T070000Z
DTEND:20250821T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/64
DESCRIPTION:Title: Convergence in law in metric and submetric spaces<
/a>\nby Prof Adam Jakubowski (Uniwersytet Mikołaja Kopernika) as part of
La Trobe University Statistics and Stochastic zoom seminar\n\n\nAbstract\n
This event is being organised in collaboration with the PVSeminar.\n\nThe
purpose of the talk is two-fold.\nFirst we will convince the audience tha
t the notion of convergence in law should not be identified with the weak
convergence of probability distributions when we leave the safe area of me
tric spaces. We shall introduce a topology related to the almost sure Skor
okhod’s representation\, which nowadays is widely used in existence prob
lems of Stochastic Partial Differential Equations. The advantage of this t
opology is its flexibility and preservation of standard methods\, proper t
o the metric case\, in the large class of so-called submetric spaces.\n\nA
topological space is submetric if there exists a continuous metric\, defi
ned on it\, and generating a metric topology (usually weaker than the orig
inal topology). As a well-known example\, an infinite dimensional\, separa
ble Hilbert space with the weak topology may serve here.\n\nThe second par
t of the talk will be devoted to the presentation of the many properties o
f submetric spaces\, which make them a suitable and intuitive tool in the
theory of stochastic processes.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
64/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Dr Andreas Søjmark (London School of Economics and Political Scie
nce)
DTSTART:20250904T070000Z
DTEND:20250904T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/65
DESCRIPTION:Title: Particle systems with moving boundaries governed b
y singular forward-backward interactions\nby Dr Andreas Søjmark (Lond
on School of Economics and Political Science) as part of La Trobe Universi
ty Statistics and Stochastic zoom seminar\n\n\nAbstract\nThis event is bei
ng organised in collaboration with the PVSeminar.\n\nThe topic of this ta
lk is an interacting system of N particles and N moving boundaries which e
volve until a final time T. If the particles hit their respective boundari
es before or at this time\, then they are absorbed\, and it is the possibi
lity of this occurring that determines how the boundaries evolve. Before t
ime T\, the boundaries should reflect both the realized impact of past abs
orptions and the expected impact of possible future absorptions\, leading
to a forward-backward specification. I will give a precise formulation of
this problem and address its well-posedness along with a suitable notion o
f Markovianity. This will take us to a new and interesting type of free bo
undary problem given by a ‘cascade’ of PDEs representing different con
figurations of absorbed particles. The motivation comes from the analysis
of contagion in financial networks which I shall briefly discuss. Based on
joint work with P Jettkant.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
65/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Maha Alghamdi (La Trobe University)
DTSTART:20251010T060000Z
DTEND:20251010T070000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/66
DESCRIPTION:Title: Multiscaling asymptotic behavior of solutions to r
andom high-order heat equations\nby Maha Alghamdi (La Trobe University
) as part of La Trobe University Statistics and Stochastic zoom seminar\n\
n\nAbstract\nThis talk focuses on high-order partial differential equation
s with random initial conditions that exhibit both long-memory and cyclic
behavior. The study considers cases where the random initial conditions po
ssess spectral singularities either at zero (corresponding to classical lo
ng-range dependence) or at nonzero frequencies (corresponding to cyclic lo
ng-range dependence). Using spectral methods and scaling techniques\, it i
s shown that\, after suitable rescaling and normalization\, the solutions
converge to Gaussian random fields. For each class of equations\, spectral
representations and covariance structures of the limiting fields are pres
ented. In the case of odd-order equations\, kernel averaging is applied to
obtain nonexplosive and nondegenerate limits. The results demonstrate tha
t the nature of the limiting fields depends on whether the equation is of
even or odd order and on the presence or absence of a spectral singularity
at zero. Several numerical examples illustrate the theoretical findings.\
n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
66/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Udara Kumaranathunga (La Trobe University)
DTSTART:20251001T020000Z
DTEND:20251001T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/67
DESCRIPTION:Title: Meta analysis of quantiles and functions of quanti
les\nby Udara Kumaranathunga (La Trobe University) as part of La Trobe
University Statistics and Stochastic zoom seminar\n\n\nAbstract\nMeta-ana
lysis of continuous outcomes has traditionally relied on methods that use
moment-based statistical measures such as means and standard deviations. H
owever\, it is often found that in practical scenarios\, outcome measures
are often skewed. Diabetes-related biomarkers\, such as creatinine\, trigl
ycerides and microalbumin are several of such examples\, which commonly di
splay underlying skewed distributions. For such skewed outcome measures\,
we introduce meta-analyses of medians\, other quantiles and functions of q
uantiles to comprehensively synthesis information. We consider the commonl
y reported scenarios of five number summary (minimum\, first quartile\, me
dian\, third quartile\, maximum) together with the sample sizes. Within a
novel density-based framework\, without making any prior assumptions about
the underlying distributions\, we use flexible quantile-based distributio
ns with percentile matching to estimate the unknown parameters. Additional
ly\, we extend the quantile estimation method to meta-analyse quantiles. W
e carry out simulation studies to evaluate the approaches and present resu
lts for varying number of studies\, study sizes\, distributions of the out
come measures\, and heterogeneity. Together with the simulation results\,
we report results from an empirical study of meta-analysis of quantiles an
d interquartile range widths as an example of meta-analysis of functions o
f quantiles\, using real-world data from the Australian Diabetes\, Obesity
\, and Lifestyle (AusDiab) study. We further present our CRAN-based R pa
ckage (metaquant) that contains the code necessary for straightforward imp
lementation of the introduced approaches.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
67/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Masters students (La Trobe University)
DTSTART:20251030T010000Z
DTEND:20251030T020000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/68
DESCRIPTION:Title: Data Science Master's final theses presentations<
/a>\nby Masters students (La Trobe University) as part of La Trobe Univers
ity Statistics and Stochastic zoom seminar\n\n\nAbstract\n12:00 Aarjav Kha
ra. Feature Selection and Scaling Directions for Robust Property Price Pre
diction\n\n12:25 Helen Arnold. A comparison of heterogeneity estimators in
meta-analysis: an online tool for researchers\, educators\, and students\
n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
68/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Krishna Suppiah\, David Teakle (La Trobe University)
DTSTART:20251113T010000Z
DTEND:20251113T020000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/69
DESCRIPTION:Title: Honour's and Master's final theses presentations<
/a>\nby Krishna Suppiah\, David Teakle (La Trobe University) as part of La
Trobe University Statistics and Stochastic zoom seminar\n\n\nAbstract\n12
:30 Krishna Suppiah. Study of Gegenbauer Time Series and their Statistics\
n\n12:55 David Teakle. The Generalized Adaptive Signed Correlation Index\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
69/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Various (Various)
DTSTART:20251218T010000Z
DTEND:20251218T030000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/70
DESCRIPTION:Title: Extended La Trobe University Statistics Seminar\nby Various (Various) as part of La Trobe University Statistics and Stoc
hastic zoom seminar\n\n\nAbstract\nExtended La Trobe University Statistics
Seminar\nHonouring the Retirement of A/Professors Paul Kabaila\nMonday 15
December\, 12:00\, PS2-313\n\nSpeakers:\n\n• Professor Alan H. Welsh\n
• Dr J. H. D. S. P. Tissera\n• Professor Chris J. Lloyd\n• Dr Waruni
Abeysekera\n• Dr Rheanna Mainzer\n• Dr Khageswor Giri\n• Dr Ayesha
Perera\n• Dr Rupert Kuveke\n• Dr Christeen Wijethunga\n• Professor S
idney Morris\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
70/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Serhii Kravchenko (La Trobe University)
DTSTART:20251215T050000Z
DTEND:20251215T060000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/71
DESCRIPTION:Title: Inference for Stochastic Processes from Discretely
Sampled Data\nby Serhii Kravchenko (La Trobe University) as part of L
a Trobe University Statistics and Stochastic zoom seminar\n\n\nAbstract\nP
hD progress talk.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
71/
END:VEVENT
BEGIN:VEVENT
SUMMARY:PhD students (La Trobe University)
DTSTART:20251127T010000Z
DTEND:20251127T010000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/72
DESCRIPTION:Title: PhD students' preconference talks\nby PhD stud
ents (La Trobe University) as part of La Trobe University Statistics and S
tochastic zoom seminar\n\n\nAbstract\nPhD students' preconference talks fo
r AustMS 2025 and ASC 2025.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
72/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Nemini Samarakoon (La Trobe University)
DTSTART:20251218T070000Z
DTEND:20251218T080000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/73
DESCRIPTION:Title: Construction\, Properties and Simulation of Haar-B
ased Multifractional Processes with Rmfrac R package\nby Nemini Samara
koon (La Trobe University) as part of La Trobe University Statistics and S
tochastic zoom seminar\n\n\nAbstract\nMultifractional processes are non-st
ationary stochastic processes that allow modeling the time-dependent regul
arity in their trajectories. This talk introduces a new class of multifrac
tional processes\, Gaussian Haar-based multifractional processes (GHBMP) b
ased on their Haar wavelet series representation. GHBMP provides computati
onally efficient simulation algorithms and can capture abrupt changes in t
he roughness of trajectories. The theoretical properties of the GHBMP will
also be discussed. Additionally\, Rmfrac R package will be introduced. Rm
frac facilitates the simulation and analyses of multifractional processes.
\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
73/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Prof O.Aryasova (Friedrich Schiller University\, Jena\, Germany)
DTSTART:20260319T010000Z
DTEND:20260319T020000Z
DTSTAMP:20260404T095357Z
UID:StatisticsandStochastic/74
DESCRIPTION:Title: The hard membrane process and transport barriers o
f turbulent flows\nby Prof O.Aryasova (Friedrich Schiller University\,
Jena\, Germany) as part of La Trobe University Statistics and Stochastic
zoom seminar\n\n\nAbstract\nIn confined fusion plasma devices like tokamak
s\, turbulence is always present to some degree and it has the bad effect
of dispersing heat and particles from the central very hot and dense regio
n to the boundary and then to the surrounding areas and walls. Sometimes b
arriers arise in intermediate regions\, which reduce this dispersion. Thes
e barriers\, also called zonal flows\, are thin layers of plasma where the
fluid velocity is not turbulent as everywhere else but ordered\, roughly
laminar\, directed perpendicularly to the direction of dispersion.\n\nTher
e is still active research to understand how these barriers arise\, whethe
r they can be triggered\, how they evolve\, the precise links with other p
lasma dynamical features and parameters. Our aim here is to rigorously def
ine a mathematical model of a sharp heat-diffusion barrier. In a scaling l
imit of theturbulence model with separation of scales we get a heat equati
on with space-dependent diffusion coefficient\, poorly diffusing near the
barrier\; then we investigate the scaling limit when the diffused barrier
converges to a sharp separating surface and describe the limit by means of
the stochastic process called Brownian motion with a hard membrane.\n
LOCATION:https://stable.researchseminars.org/talk/StatisticsandStochastic/
74/
END:VEVENT
END:VCALENDAR