Statistics Across Campuses

  On arbitrarily underdispersed discrete distributions Date: 28 April 2022, Thursday Time: 12pm AEDT Contact the organizer: Andriy Olenko a.olenko@latrobe.edu.au Speaker: Dr Alan Huang, UQ Abstract: We review a range of generalized count distributions, investigating which (if any) can be arbitrarily underdispersed, i.e., its variance can be arbitrarily small compared to its mean. A philosophical implication is that models failing this criterion perhaps should not be considered a “statistical model” according to the extendibility criterion of McCullagh (2002). Four practical implications will be discussed. We suggest that all generalizations of the Poisson distribution be tested against this property. Zoom meeting link: Join from a PC, Mac, iOS or Android: https://latrobe.zoom.us/j/98357628534 Or iPhone one-tap (Australia Toll):  +61280152088,98357628534# Or Telephone:     Dial: +61 2 8015 2088     Meeting ID: 983 5762 8534     Int...

Twisted: Improving particle filters by learning modified paths Date: 22 April 2022, Friday Time: 4pm AEDT Speaker: Dr Joshua Bon (QUT) Abstract: Particle filters, and sequential Monte Carlo (SMC) more generally, operate by propagating weighted samples (or particles) through a sequence of distributions. Such a sequence is characterised by a Feynman-Kac model (or path measure) and chosen for the given inferential task at hand. One can also define twisted Feynman-Kac models which preserve the inferential target but provide a more efficient sequence of distributions (or path) for the SMC algorithm to use. Optimally twisted models define perfect Monte Carlo samplers and are therefore an important concept for SMC algorithms. We investigate how to learn and use twisted Feynman-Kac models in situations where the original model involves difficult or intractable transition dynamics. This extends existing work which relies on twisting the model analytically. We achieve twisting via Monte Carlo a...

Statistical analysis of machine learning methods Date :  Thursday, 14 April 2022 Time:   4 pm   - 5 pm Speaker:    Professor Johannes Schmidt-Hieber Abstract: Recently a lot of progress has been made regarding the theoretical understanding of machine learning methods. One of the very promising directions is the statistical approach, which interprets machine learning as a collection of statistical methods and builds on existing techniques in mathematical statistics to derive theoretical error bounds and to understand phenomena such as overparametrization. The talk surveys this field and describes future challenges. Zoom Link:  Please contact Yanrong Yang ( yanrong.yang@anu.edu.au ) to obtain the zoom link for this seminar.

  Statistical Challenges in Stellar Parameter Estimation from Theory and Data. Date: Tuesday, 12 April 2022   Time: 16:00 (UTC) 3:00 (AEDT)   Speaker:  Josh Speagle (Toronto University, Canada) Contact the organizer: Andriy Olenko a.olenko@latrobe.edu.au Abstract:  Understanding how the Milky Way fits into the broader galaxy population requires studying the properties of other galaxies as well as our own. While it is possible to observe the structure of other galaxies directly, understanding the structure of our own Galaxy from within requires inferring the 3-D positions, velocities, and other properties of billions of stars. In this talk, I will discuss some of the statistical challenges in inferring stellar parameters from modern photometric surveys such as Gaia and SDSS, focusing in particular on issues with existing theoretical stellar models, the complex nature of parameter uncertainties, and scalability to large datasets. I will then describe so...

                 Online Estimation for Functional Data Date :  Thursday, 7 April 2022 Time:   11 am  - 12 noon Speaker:  Professor Fang Yao Abstract: Functional data analysis has attracted considerable interest, and is facing new challenges of the increasingly available data in streaming manner. In this work, we propose a new online method to dynamically update the local linear estimates of mean and covariance functions of functional data, which is the foundation of subsequent analysis. The kernel-type estimates can be decomposed into two sufficient statistics depending on the data-driven bandwidths. We propose to approximate the future optimal bandwidths by a dynamic sequence of candidates and combine the corresponding statistics across blocks to make an updated estimation. The proposed online method is easy to compute based on the stored sufficient statistics and current data block. Based on the asymptotic normality of t...

Detection boundaries for sparse gamma scale mixture models Time: 2-3PM 8 April  Location: Carslaw 829 (University of Sydney) or Zoom at https://uni-sydney.zoom.us/j/87417817957  Speaker: Michael Stewart - University of Sydney Mixtures of distributions from a parametric family are useful for various statistical problems, including nonparametric density estimation, as well as model-based clustering. In clustering an enduringly difficult problem is choosing the number of clusters; when using mixtures models for model-based clustering this corresponds (roughly) to choosing the number of components in the mixture. The simplest version of this model selection problem is choosing between a known single-component mixture, and a "contaminated" version where a second unknown component is added. Due to certain structural irregularities, many standard asymptotic results from hypothesis testing do not apply in these "mixture detection" problems, including those relating to p...

                Recent Development of Rank-Constrained and Distributed Statistical Learning Date :  Thursday, 31 March 2022 Time:   10 am  - 11 am Speaker:  Assistant Professor  Ziwei Zhu Abstract: In this talk, I will present two recent works on rank-constrained least squares and distributed statistical learning respectively. The first part of the talk highlights a near optimal in-sample prediction error bound for the rank-constrained least squares estimator with no assumption on the design matrix. Lying at the heart of the proof is a covering number bound for the family of projection operators corresponding to the subspaces spanned by the design. By leveraging this complexity result, we perform a power analysis for a permutation test on the existence of a low-rank signal under the high-dimensional trace regression model. The second part of the talk proposes a new one-shot distributed learning algorithm through refitting ...

Power Analysis for Cluster Randomized Trials with Multiple Primary Endpoints Date: 1 April, Friday Time: 2pm AEDT Speaker :  Professor Song Zhang   (UT Southwestern Medical Center) Cluster randomized trials (CRTs) are widely used in different areas of medicine and public health. Recently, with the increasing complexity of medical therapies and technological advances in monitoring multiple outcomes, many clinical trials attempt to evaluate multiple primary endpoints. In this study we present a power analysis method for CRTs with K > 2 binary co-primary endpoints. It is developed based on the GEE (generalized estimating equation) approach, and three types of correlations are considered: inter-subject correlation within each endpoint, intra-subject correlation across endpoints, and inter-subject correlation across endpoints. A closed-form joint distribution of the K test statistics is derived, which facilitates the evaluation of power and type I error for arbitrarily construct...

Score matching for microbiome compositional data Date: 25 March 2022, Friday Time: 4pm AEDT Speaker: A/Prof Janice Scealy (ANU) Abstract: Compositional data are challenging to analyse due to the non-negativity and sum-to-one constraints on the sample space. It is often the case with microbiome compositional data that many of the components are highly right-skewed, with large numbers of zeros. A major limitation of currently available estimators for compositional models is that they either cannot handle many zeros in the data or are not computationally feasible in moderate to high dimensions. We derive a new set of novel score matching estimators applicable to distributions on a Riemannian manifold with boundary, of which the standard simplex is a special case. The score matching method is applied to estimate the parameters in a new flexible model for compositional data and we show that the estimators are scalable and available in closed form. We apply the new model and estimators to r...

               Factor modelling for high-dimensional functional time series Date :  Thursday, 17 March 2022 Time:   9  am  - 10 am Speaker:  Assistant Professor Xinghao Qiao Abstract: Many economic and scientific problems involve the analysis of high-dimensional functional time series, where the number of functional variables (p) diverges as the number of serially dependent observations (n) increases. In this talk, we present a novel functional factor model for high-dimensional functional time series that maintains and makes use of the functional and dynamic structure to achieve great dimension reduction and find the latent factor structure. To estimate the number of functional factors and the factor loadings, we propose a fully functional estimation procedure based on an eigenanalysis for a nonnegative definite matrix. Our proposal involves a weight matrix to improve the estimation efficiency and tackle the issue of heter...