Statistics Across Campuses

Recognising research software in academia Date: 29 October 2021, Friday Time: 4pm AEDT Speaker: Dr Nicholas Tierney (Telethon Kids Institute) Abstract: We need software to do our research. Many researchers write software which implements new methodologies, statistical models, or graphics. Without software, research would not have nearly the same impact or utility. Despite the obvious impact of software, we are still working out how to adequately acknowledge research software in academia, and how to provide rewarding career paths for those who want to write research software as academic output. The relatively new field of research software engineering can help address this. A research software engineer combines professional software expertise with an understanding of research. I work as a research software engineer in academia, and I wanted to explain how this role fits into academia. Specifically, I'll discuss why we need to consider software as academic output, what a ...

         Estimation of the Distribution of Episodically Consumed Foods Measured with Error Date :  Thursday, 21 October 2021 Time:   10 am  - 11 am Speaker:  Professor Aurore Delaigle Abstract: Dietary data collected from 24-hour dietary recalls are observed with significant measurement errors. In the nonparametric curve estimation literature, a lot of effort has been devoted to designing methods that are consistent under contamination by noise. However, some foods such as alcohol or fruits are consumed only episodically, and may not be consumed during the day when the 24-hour recall is administered. Existing nonparametric methods cannot deal with those so-called excess. We present new semiparametric estimators of the distribution of such episodically consumed food data. Zoom Link:  Please contact Yanrong Yang ( yanrong.yang@anu.edu.au ) to obtain the zoom link for this seminar.

Do you have a moment? Bayesian inference using estimating equations via empirical likelihood Date: 22 October 2021, Friday Time: 4pm AEDT Speaker: Prof Howard Bondell (University of Melbourne) Abstract: Bayesian inference typically relies on specification of a likelihood as a key ingredient. Recently, likelihood-free approaches have become popular to avoid specification of potentially intractable likelihoods. Alternatively, in the Frequentist context, estimating equations are a popular choice for inference corresponding to an assumption on a set of moments (or expectations) of the underlying distribution, rather than its exact form. Common examples are in the use of generalised estimating equations for marginal estimation of correlated responses, or in the use of M-estimators for robust regression avoiding the distributional assumptions on the errors. In this talk, I will discuss some of the motivation behind empirical likelihood, and how it can be used to incorporate a fully Bayesia...

       Gaussian approximation for high-dimensional data: Recent progress Date :  Thursday, 21 October 2021 Time:   4-5 pm Speaker:  Prof Yuta Koike (University of Tokyo) Abstract: We review recent progress in Gaussian approximation of a sum of high-dimensional independent random vectors (and related statistics) on hyper-rectangles, where the dimension can be much larger than the sample size. Such approximation is useful for justifying bootstrap approximation of maximum statistics in high-dimensional data and is therefore important for uniform inference in high-dimensional models. Zoom Link:  https://macquarie.zoom.us/j/88030361110?pwd=OTAzRStXNEtmTGRyaWRKcVJLL09Pdz09

Spatio-temporal joint species distribution modeling – A community-level basis function approach Date: 21 October 2021, Thursday Time: 12pm AEST Contact the organizer: Andriy Olenko a.olenko@latrobe.edu.au Speaker: Dr Francis K.C. Hui, ANU Abstract: The last decade in ecology has seen the development and rising popularity of joint species distribution modeling approaches for studying species assemblages, with by far the most common approach being based around generalized linear latent variable models (LVMs). However, while methodological and computational advances continue to be made with LVMs, their application to spatio-temporal multivariate abundance data i.e., observations of multiple species recorded across space and/or time, remains computationally challenging and not necessarily scalable when it comes to fitting and inference. In this talk, we propose an alternative approach to spatio-temporal joint species distribution modeling which breaks away from the LVM framework. Inspired...

        Predicting Returns with Text Data Date :  Thursday, 14 October 2021 Time:   10 am  - 11 am Speaker: Prof Dacheng Xiu Abstract: We introduce a new text-mining methodology that extracts information from news articles to predict asset returns. Unlike more common sentiment scores used for stock return prediction (e.g., those sold by commercial vendors or built with dictionary-based methods), our supervised learning framework constructs a score that is specifically adapted to the problem of return prediction. Our method proceeds in three steps: 1) isolating a list of terms via predictive screening, 2) assigning prediction weights to these words via topic modeling, and 3) aggregating terms into an article-level predictive score via penalized likelihood. We derive theoretical guarantees on the accuracy of estimates from our model with minimal assumptions. In our empirical analysis, we study one of the most actively monitored streams of news articles i...

       Nonparametric Estimation of Repeated Densities with Heterogeneous Sample Sizes Date :  Thursday, 7 October 2021 Time:   10 am  - 11 am Speaker:  Dr Xiongtao Dai   Abstract: Functional data analysis concerns a sample of random functions, such as a collection of body growth trajectories. Dimension reduction tools, such as functional principal component analysis, are available to reduce and represent the infinite-dimensional functions. In this work, we are interested in estimating densities as functions, where each density comes from a subpopulation. For example, in the context of epidemiology, the age distributions of patients with different diseases is of central interest, where each disease defines a subpopulation. A key challenge comes from the highly variable sample sizes for different conditions, making the estimation of age profiles difficult for rare conditions. We propose a fully data-driven approach to estimate the densities wit...

The International Astrostatistics Association and the IAU Astroinformatics and Astrostatistics Commission are starting a new international online seminars series on Statistics and Data Science applications in Astronomy . The seminar will alternate between Europe-US and Australasia-Europe on a monthly basis. The first seminar: Adventures in Astronomical Time Series Analysis Date: 12 October 2021, Tuesday Time: 4pm UTC Contact the organizer: Andriy Olenko a.olenko@latrobe.edu.au Speaker: Jeffrey D. Scargle (NASA Ames Research Center, US) Abstract: Welcome to a tour of the volatile, highly active Universe — in stark contrast to earlier serene '"clockwork’’ visions. Innovative data analysis techniques have illuminated explosive physical processes animating these systems. Examples include a Fourier transform suited to the irregular sampling characteristic of much astronomical data, but time domain techniques will be emphasized for these applications: gamma-ray activity in the C...