Robert Kohn | ARC Centre of Excellence for Mathematical and Statistical Frontiers

UNSW Sydney

Robert Kohn is a Scientia Professor in the School of Economics at UNSW School of Business. He is a former Head of the Statistics and Operations Group at UNSW Australian Graduate School of Management, and previously held an Australian Research Council Discovery Outstanding Researcher Award (DORA).

Robert has pioneered Bayesian methodology in time series, nonparametric regression, graphical models, variable selection, covariance selection and copula models. His primary current research interests are in Bayesian inference and Bayesian computation, in particular for models with intractable likelihoods, and the application of the methodology to a number of model classes. His areas of application are health and financial econometrics.

Robert is a Fellow of the Academy of the Social Sciences in Australia, Fellow of the American Statistical Association, Fellow of the Institute of Mathematical Statistics, and Fellow of the Journal of Econometrics. He has over 8,000 Google scholar citations and an h-index of 42. He has had various editorial roles and is currently Associate Editor of Bayesian Analysis and the Journal of Econometrics.

Research Interests:

Bayesian methodolgy

Bayesian nonparametric methods

Markov chain Monte Carlo simulation algorithms

Model averaging

Multivariate Gaussian and non-Gaussian regression

Nonparametric regression

Time series modelling

Variable selection

Projects

Developing flexible copula based methods

A copula is a multivariate model whose marginals are uniform. A copula based model is one that is based on a copula. The attraction of using copula models is that we can separately model the marginal distributions and the dependence structure of our target distribution. Such models are particularly attractive when some or all of the marginals are discrete or a mixture of discrete and continuous components. A multivariate probit model is one simple example of a copula model.

Enabling Bayesian inference for big data

In the last decade or so, there has been a dramatic increase in storage facilities and the possibility of processing huge amounts of data. This has made large high-quality data sets widely accessible for practitioners. This technology innovation seriously challenges inference methodology, in particular simulation algorithms commonly applied in Bayesian inference. These algorithms typically require repeated evaluations over the whole data set when fitting models, precluding their use in the age of so called big data.

Flexible Bayesian Methods for time series and longitudinal panel data models

Many statistical applications use models that incorporate latent variables. For example, random effect panel data models, use latent variables to account for dependence between observations. State space models whose latent variables follow a Markov process, are used in economics, finance, and engineering.

Flexible models and methods for cognitive model-based decision making

This project pursues breakthroughs which allow important questions of basic and applied science to be addressed using mathematical decision models. Advances are made through an interdisciplinary effort, combining recent developments in econometric and statistical methods, cognitive science, and advances in computing. The outcomes will bring to a new range of questions a proven and powerful approach for investigating psychological effects. The project will begin this expansion effort with investigations of choices about health care and consumer purchases.

Hamiltonian Monte Carlo with Energy Conserving Subsampling

ACEMS researchers found a way to make Hamiltonian Monte Carlo (HMC) compatible with data subsampling without deteriorating the sampling efficiency of an algorithm.

Particle Methods for Stochastic Differential Equation Mixed Effects Models

ACEMS Researchers at QUT and UNSW introduce three novel parameter estimation methods for stochastic differential equation mixed effects models (SDEMEMs).

Variance reduction properties of the reparameterization trick

ACEMS researchers have now provided a mathematical treatment that sheds light on the variance reduction properties of the reparameterization trick. The work was presented at the 22nd International Conference on Artificial Intelligence and Statistics (AISTATS) in Okinawa, Japan in April 2019, and was subsequently published in the conference proceedings. AISTATS is considered a top conference in Machine Learning.

Publications

Invited talks, refereed proceedings and other conference outputs

2020

Botha, I., Drovandi C., South L., & Kohn R. (2020). Tuning the number of state particles in exact-approximate SMC. Bayesian Young Statisticians Meeting: Online (BAYSM:O).

2019

Nguyen, N., Tran M-N., Kohn R., & Nott D. (2019). Stochastic Variational Bayes with Particle Filter. The 12th International Conference on Monte Carlo Methods and Applications.

Botha, I., Kohn R., & Drovandi C. (2019). Particle Methods for Stochastic Differential Equation Mixed Effects Models. Bayes on the Beach 2019.

Xu, M.., Quiroz M., Kohn R., & Sisson S. (2019). Variance reduction properties of the reparameterisation trick. (Chaudhuri, K., & Sugiyama M., Ed.).The 22nd International Conference on Artificial Intelligence and Statistics. 89, 2711-2720.

Nguyen, N., Gunawan D., Tran M-N., & Kohn R. (2019). Long short term memory stochastic volatility. The 3rd International Conference on Econometrics and Statistics (EcoSta 2019).

Botha, I., Kohn R., & Drovandi C. (2019). Bayesian Parameter Inference for Stochastic Differential Equation Mixed Effects Models. The 12th International Conference on Monte Carlo Methods and Applications.

Gunawan, D., Dang K-D., Quiroz M., Kohn R., & Tran M-N. (2019). Subsampling Sequential Monte Carlo for Static Bayesian Models. The 12th International Conference on Monte Carlo Methods and Applications.

Dang, K-D., Quiroz M., Kohn R., Tran M-N., & Villani M. (2019). Efficient Bayesian inference for large data sets by HMC with energy conserving subsampling. The 10th European Seminar on Bayesian Econometrics.

2018

Dang, K.. D., Quiroz M., Kohn R., Tran M N., & Villani M. (2018). Hamiltonian Monte Carlo with energy conserving subsampling. The 2nd International Conference on Econometrics and Statistics (EcoSta 2018).

Quiroz, M., Nott D., & Kohn R. (2018). Gaussian variational approximation for high-dimensional state space models. 2nd International Conference on Econometrics and Statistics.

Quiroz, M., Tran M N., Villani M., Kohn R., & Dang K-D. (2018). The block-Poisson estimator for optimally exact subsampling MCMC. International Society of Bayesian Analysis.

Gunawan, D.., Carter C.., & Kohn R. (2018). Efficiently combining Pseudo Marginal and Particle Gibbs sampling. Joint Statistical Meeting 2018.

Quiroz, M., Nott D., & Kohn R. (2018). The Mathematics of Biological Systems Management Symposium. The Mathematics of Biological Systems Management Symposium.

2017

Dang, K.-D., Quiroz M., Kohn R., Tran M N., & Villani M. (2017). Hamiltonian Monte Carlo with Energy Conserving Subsampling. Bayes on the Beach.

Tran, M N., Kohn R., Quiroz M., & Villani M. (2017). The block-pseudo marginal sampler. The 1st International Conference on Econometrics and Statistics (EcoSta 2017).

Tran, M N., Kohn R., Quiroz M., & Villani M. (2017). The block pseudo-marginal sampler. Bayes on the beach.

Quiroz, M., Nott D., & Kohn R. (2017). Gaussian variational approximation for high-dimensional state space models. The Sydney Time Series & Forecasting Symposium.

Quiroz, M., Tran M N., Villani M., & Kohn R. (2017). Exact Subsampling MCMC. 1st International Conference on Econometrics and Statistics (EcoSta 2017).

2016

Kohn, R. (2016). Speeding Up MCMC by Efficient Data Subsampling. ISBA 2016 World Meeting, Sardinia .

Journal Articles

2021

Botha, I., Kohn R., & Drovandi C. (2021). Particle Methods for Stochastic Differential Equation Mixed Effects Models. Bayesian Analysis. 16(2), 575-609. doi: 10.1214/20-BA1216

2020

Chin, V.., Gunawan D.., Fiebig D.., Kohn R., & Sisson S. (2020). Efficient data augmentation for multivariate probit models with panel data: An application to general practitioner decision-making about contraceptives. Journal of Royal Statistical Society. 69(2), 277-300. doi: 10.1111/rssc.12393

Balnozan, I., Fiebig D. G., Asher A., Kohn R., & Sisson S. (2020). Hidden Group Time Profiles: Heterogeneous Drawdown Behaviours in Retirement. arXiv. arXiv:2009.01505v1,

Tran, M-N., Nguyen N., Nott D., & Kohn R. (2020). Bayesian Deep Net GLM and GLMM. Journal of Computational and Graphical Statistics. 29, 97-113. doi: 10.1080/10618600.2019.1637747

Mendes, E.. F., Carter C.. K., Gunawan D.., & Kohn R. (2020). A Flexible particle Markov chain Monte Carlo Method. Statistics and Computing. 30, 783-798. doi: 10.1007/s11222-019-09916-7

2019

Chin, V., Gunawan D., Fiebig D. G., Kohn R., & Sisson S. (2019). Efficient data augmentation for multivariate probit models with panel data: An application to general practitioner decision-making about contraceptives. arXiv. arXiv:1806.07274v2.

Gunawan, D.., Tran M.-N.., Suzuki K.., Dick J.., & Kohn R. (2019). Computationally efficient Bayesian estimation of high-dimensional Archimedean copulas with discrete and mixed margins. Statistics and Computing. 29(5), 933-946. doi: 10.1007/s11222-018-9846-y

Dang, K-D., Quiroz M., Kohn R., Tran M-N., & Villani M. (2019). Hamiltonian Monte Carlo with Energy Conserving Subsampling. Journal of Machine Learning Research. 20(100), 1-31.

Frazier, D. T., Nott D. J., Drovandi C., & Kohn R. (2019). Bayesian inference using synthetic likelihood: asymptotics and adjustments. arXiv. arXiv:1902.04827v2.

Salomone, R., Quiroz M., Kohn R., Villani M., & Tran M-N. (2019). Spectral Subsampling MCMC for Stationary Time Series. arXiv. arXiv:1910.13627v1.

Quiroz, M., Kohn R., Villani M., & Tran M N. (2019). Speeding Up MCMC by Efficient Data Subsampling. Journal of the American Statistical Association. 114(526), 831-843. doi: 10.1080/01621459.2018.1448827

2018

Quiroz, M., Nott D. J., & Kohn R. (2018). Gaussian variational approximations for high-dimensional state space models.

Quiroz, M., Villani M., Kohn R., Tran M-N., & Dang K-D. (2018). Subsampling MCMC - an Introduction for the Survey Statistician. Sankhya A. 80(1), 33-69. doi: 10.1007/s13171-018-0153-7

Quiroz, M., Tran M-N., Villani M., & Kohn R. (2018). Speeding up MCMC by Delayed Acceptance and Data Subsampling. Journal of Computational and Graphical Statistics. 27(1), 12 - 22. doi: 10.1080/10618600.2017.1307117

2017

Khaled, M., & Kohn R. (2017). The approximation properties of copulas by mixtures.

Gunawan, D., Carter C., & Kohn R. (2017). Efficient Bayesian inference for multivariate factor stochastic volatility models with leverage.

Tran, M N., Nott D. J., & Kohn R. (2017). Variational Bayes With Intractable Likelihood. Journal of Computational and Graphical Statistics. 26(4), 873 - 882. doi: 10.1080/10618600.2017.1330205

Khaled, M., & Kohn R. (2017). The approximation properties of copulas by mixtures.

Gunawan, D., Carter C., Fiebig D. G., & Kohn R. (2017). Efficient Bayesian Estimation for Flexible Panel Models for Multivariate Outcomes: Impact of Life events on mental health and excessive alcohol consumption.

Gunawan, D., Carter C., Fiebig D. G., & Kohn R. (2017). Efficient Bayesian Estimation for Flexible Panel Models for Multivariate Outcomes: Impact of life events on mental health and excessive alcohol consumption.

Dang, K.. D., Quiroz M., Kohn R., Tran M N., & Villani M. (2017). Hamiltonian Monte Carlo with Energy Conserving Subsampling.

Gunawan, D., Khaled M., & Kohn R. (2017). Mixed Marginal Copula Modeling.

2016

Tran, M N., Nott D., Kuk A., & Kohn R. (2016). Parallel variational Bayes for large datasets with an application to generalized linear mixed models. Journal of Computational and Graphical Statistics. 25(2), 626-646. doi: 10.1080/10618600.2015.1012293

Tran, M N., Pitt M. K., & Kohn R. (2016). Adaptive Metropolis–Hastings sampling using reversible dependent mixture proposals. Statistics and Computing. 26(1-2), 361-381. doi: 10.1007/s11222-014-9509-6

Del Moral, P., Kohn R., & Patras F. (2016). On particle Gibbs samplers. Annales de l'IHP-Probabilités et Statistiques. 52(4), 1687-1733. doi: 10.1214/15-AIHP695

Gunawan, D., Tran M N., Suzuki K.., Dick J.., & Kohn R. (2016). Computationally efficient Bayesian estimation of high dimensional copulas with discrete and mixed margins.

Scharth, M., & Kohn R. (2016). Particle efficient importance sampling. Journal of Econometrics. 190(1), 133-147. doi: 10.1016/j.jeconom.2015.03.047

2015

Doucet, A., Pitt M. K., Deligiannidis G., & Kohn R. (2015). Efficient implementation of Markov chain Monte Carlo when using an unbiased likelihood estimator. Biometrika. 102(2), 295-313. doi: 10.1093/biomet/asu075

Del Moral, P., Kohn R., & Patras F. (2015). A duality formula for Feynman–Kac path particle models. Comptes Rendus Mathematique. 353(5), 465-469. doi: 10.1016/j.crma.2015.02.008

2014

Peters, G. W., Dong A. X. D., & Kohn R. (2014). A copula based Bayesian approach for paid–incurred claims models for non-life insurance reserving. Insurance: Mathematics and Economics. 59, 258-278. doi: 10.1016/j.insmatheco.2014.09.011

Tran, M N., Giordani P., Mun X., Kohn R., & Pitt M. K. (2014). Copula-Type Estimators for Flexible Multivariate Density Modeling Using Mixtures. Journal of Computational and Graphical Statistics. 23(4), 1163-1178. doi: 10.1080/10618600.2013.842918

Technical reports and unrefereed outputs

2019

Ryan, L. M., Chen V., Beavan A., Speilmann J., Sisson S., & Kohn R. (2019). A longitudinal analysis of the developmental trajectories of domain specific and domain generic abilities in high-level football players.

2018

Gunawan, D.., Carter C.., & Kohn R. (2018). Efficiently combining Pseudo Marginal and Particle Gibbs sampling.

Chin, V.., Gunawan D.., Kohn R., & Sisson S. (2018). Efficient data augmentation for multivariate probit models with panel data: An application to general practitioner decision-making about contraceptives.

Gunawan, D.., Kohn R., Carter C.., & Tran M.. N. (2018). Flexible density tempering approaches for state space models with an application to factor stochastic volatility models.

Gunawan, D.., Kohn R., Quiroz M.., Dang K.. D., & Tran M.. N. (2018). Subsampling Sequential Monte Carlo for static Bayesian Models.

Gunawan, D.., Brown S.., Kohn R., & Tran M.. N. (2018). New Estimation Approaches for the Linear Ballistic Accumulator Model.

2017

Gunawan, D., Carter C., & Kohn R. (2017). Efficient Bayesian inference for multivariate factor stochastic volatility models with leverage.

Dang, K.-D., Quiroz M., Kohn R., Tran M N., & Villani M. (2017). Hamiltonian Monte Carlo with Energy Conserving Subsampling.

2016

Tran, M N., Kohn R., Quiroz M., & Villani M. (2016). Block-Wise Pseudo-Marginal Metropolis-Hastings.

Quiroz, M., Villani M., & Kohn R. (2016). Exact Subsampling MCMC.

Gunawan, D., Tran M N., & Kohn R. (2016). Fast Inference for Intractable Likelihood Problems using Variational Bayes.

Gunawan, D., Khaled M., & Kohn R. (2016). Mixed Marginal Copula Modeling.

Gunawan, D., Tran M N., Suzuki K., Dick J., & Kohn R. (2016). Computationally Efficient Bayesian Estimation of High Dimensional Copulas with Discrete and Mixed Margins.

Research links

ORCID

Contact

r.kohn@unsw.edu.au

Welcome to the ACEMS Reportal

All contributors should enter data on all of the Centre’s reportable key performance indicators (KPIs).

Professor Robert Kohn

Projects

Developing flexible copula based methods

Enabling Bayesian inference for big data

Flexible Bayesian Methods for time series and longitudinal panel data models

Flexible models and methods for cognitive model-based decision making

Hamiltonian Monte Carlo with Energy Conserving Subsampling

Particle Methods for Stochastic Differential Equation Mixed Effects Models

Variance reduction properties of the reparameterization trick

Publications

2020

2019

2018

2017

2016

2021

2020

2019

2018

2017

2016

2015

2014

2019

2018

2017

2016

Research links

Contact

Links

ARC Centre of Excellence for Mathematical & Statistical Frontiers

Welcome to the ACEMS Reportal

All contributors should enter data on all of the Centre’s reportable key performance indicators (KPIs).

Professor Robert Kohn

Subscribe to our newsletter

Stay connected

Links

ARC Centre of Excellence for Mathematical & Statistical Frontiers