My name’s Patrick Laub, I’m a Lecturer at UNSW in the School of Risk and Actuarial Studies. I teach courses on deep learning and on statistical machine learning for actuaries. My PhD was in applied probability with a focus on computational methods, and it was jointly conducted between Aarhus University and University of Queensland. I was lucky to have as supervisors Søren Asmussen and Phil Pollett. I’m interested in the intersection of maths/stats and computing in actuarial science.
I have worked at the University of Melbourne, at ISFA in the Université Claude Bernard Lyon 1 (Lyon, France), at Google (Sydney) & Data61 (then called National ICT Australia). I have taught programming and probability at universities since 2009. For more background you can check out my LinkedIn profile or my publications below.
Software Packages
- Creator of the Python hawkesbook package to accompany our book on Hawkes processes
- Creator of the Python approxbayescomp & R approxbayescomp packages for efficient Approximate Bayesian Computation
- Creator of the Julia package EMpht.jl which fits phase-type distributions
- Creator of the Python fastEDM & R fastEDM packages, and maintainer of the Stata edm package for Empirical Dynamical Modelling
- Contributor to the Python drn package for distributional regression using neural networks
Papers
- Benjamin Avanzi, Eric Dong, Patrick J. Laub, Bernard Wong (2024), Distributional Refinement Network: Distributional Forecasting via Deep Learning, submitted [arxiv, code, package]
- Francesco Ungolo & Patrick J. Laub (2024), An Augmented Variable Dirichlet Process Mixture model for the analysis of dependent lifetimes, ASTIN Bulletin (to appear) [ssrn, code]
- Patrick J. Laub, Young Lee, Philip K. Pollett, Thomas Taimre (2024), Hawkes Models and Their Applications, Annual Review of Statistics and Its Application, accepted for 2025 issue [arxiv]
- Young Lee, Patrick J. Laub, Thomas Taimre, Hongbiao Zhao, Jiancang Zhuang (2022), Exact simulation of extrinsic stress-release processes, Journal of Applied Probability, 59(1) [article, arxiv, code]
- Pierre-Olivier Goffard, Patrick J. Laub (2021), Approximate Bayesian Computations to fit and compare insurance loss models, Insurance: Mathematics and Economics, 100, pp. 350-371 [article, arxiv, code, package]
- Jinjing Li, Michael J. Zyphur, George Sugihara, Patrick J. Laub (2021), Beyond Linearity, Stability, and Equilibrium: The edm Package for Empirical Dynamic Modeling and Convergent Cross Mapping in Stata, Stata Journal, 21(1), pp. 220-258 [article, preprint, code, package]
- Patrick J. Laub, Nicole El Karoui, Stéphane Loisel, Yahia Salhi (2020), Quickest detection in practice in presence of seasonality: An illustration with call center data, Insurance data analytics: some case studies of advanced algorithms and applications, Economica [arxiv, book, code]
- Pierre-Olivier Goffard, Patrick J. Laub (2020), Orthogonal polynomial expansions to evaluate stop-loss premiums, Journal of Computational and Applied Mathematics [article, arxiv, code]
- Søren Asmussen, Pierre-Olivier Goffard, Patrick J. Laub (2019), Orthonormal polynomial expansions and lognormal sum densities, Risk and Stochastics: Ragnar Norberg at 70 (Mathematical Finance Economics), World Scientific [amazon, arxiv, code]
- Søren Asmussen, Jevgenijs Ivanovs, Patrick J. Laub, and Hailiang Yang (2019), Phase-type models in life insurance: fitting and valuation of equity-linked benefits, Risks, 7(1), 17 pages [article (open access), code, package]
- Patrick J. Laub, Robert Salomone, Zdravko I. Botev (2019), Monte Carlo estimation of the density of the sum of dependent random variables, Mathematics and Computers in Simulation, 161, pp. 23-31 [article, arxiv, code]
- Søren Asmussen, Enkelejd Hashorva, Patrick J. Laub, Thomas Taimre (2017), Tail asymptotics of light-tailed Weibull-like sums, Probability and Mathematical Statistics, 37(2), pp. 235–256 [article, arxiv]
- Lars Nørvang Andersen, Patrick J. Laub, Leonardo Rojas-Nandayapa (2016), Efficient simulation for dependent rare events with applications to extremes, Methodology and Computing in Applied Probability, 20(1), pp. 385–409 [article, arxiv, code]
- Patrick J. Laub, Søren Asmussen, Jens Ledet Jensen, Leonardo Rojas-Nandayapa (2015), Approximating the Laplace transform of the sum of dependent lognormals, Advances in Applied Probability, 48(A), pp. 203–215 [article, arxiv, code]
- Patrick J. Laub, Thomas Taimre, Philip K. Pollett (2015), Hawkes Processes, Technical report [arxiv]
Book & Theses
- Patrick J. Laub, Young Lee, Thomas Taimre (2022), The Elements of Hawkes Processes, Springer
- Doctor of Philosophy (Applied Probability) 2018, Computational methods for sums of random variables [pdf, tex]
- Bachelor of Science (Mathematics, Hons. I) 2014, Hawkes Processes: Simulation, Estimation, and Validation [pdf]
Teaching
I created and am the lecturer-in-charge for UNSW’s courses ACTL3143 Artificial Intelligence and Deep Learning Models for Actuarial Applications and ACTL5111 Artificial Intelligence and Deep Learning Models for Risk and Insurance [materials].
I’m also the co-lecturer for UNSW’s courses titled Statistical Machine Learning for Risk and Actuarial Applications (coded ACTL3142 & ACTL5110).
Previously, at Université Claude Bernard Lyon 1, I created and lectured short courses on Rare Event Estimation [materials, lecture recordings].
Reviewer
Peer-reviewer for Annals of Actuarial Science, Annals of Operations Research, European Actuarial Journal, Insurance: Mathematics and Economics, Journal of Computational and Graphical Statistics, Lifetime Data Analysis, Methodology and Computing in Applied Probability, and Statistics & Probability Letters.
Presentations
- Strasbourg 2023, Empirical Dynamic Modelling: Automatic Causal Inference and Forecasting, Probability Group Seminar, Université de Strasbourg [slides]
- Valencia 2023, Approximate Bayesian Computation and Insurance, PARTY Conference 2023 [slides]
- University of Sydney 2022, Empirical Dynamic Modelling: Automatic Causal Inference and Forecasting, Time Series and Forecasting Symposium [YouTube, slides]
- Online (via Zoom) 2021, Approximate Bayesian Computation in Insurance, Insurance Data Science conference [YouTube, slides]
- Melbourne (via Zoom) 2021, A Software Engineer’s Toolkit for Quantitative Research, UniMelb Quantitative Methods Network Seminar [YouTube, slides]
- Melbourne (via Zoom) 2021, Approximate Bayesian Computation in Insurance, University of Melbourne Actuarial Group & Applied Probability Group seminars [slides]
- Sydney (via Zoom) 2021, Approximate Bayesian Computation & Writing Performant Python Code, UNSW Risk and Actuarial Group Seminar [YouTube, slides]
- Paris (via Webex) 2020, Approximate Bayesian Computation and Insurance, Chaire DAMI Technical Seminar [YouTube, slides]
- Munich 2019, Phase-Type Models in Life Insurance, Insurance: Mathematics and Economics (IME 2019) [actuview]
- Lyon 2018, Phase-Type Models in Life Insurance, Institut de Science Financière et d’Assurances Séminaire Labo [YouTube]
- Brisbane 2017, Rare-event asymptotics and estimation for dependent random sums, a talk at the UQ SMORS Seminar series [slides]
- Sydney 2017, Efficient simulation for dependent rare events with applications to extremes, a UNSW Probability and Statistics seminar [slides]
- Uluru 2017, Tail asymptotics of light-tailed Weibull-like sums, a short talk at the Probability @ the Rock conference in honour of Phil Pollett. This presentation was “highly commended” [slides]