Claim reserving & pricing
Micro-level and aggregate reserving, reporting delays, IBNR estimation, population-sampling foundations, and coherent links between reserves and premiums.
Actuarial science · Insurance analytics
Risk models for a data-rich world.
I am a Professor of Actuarial Science at the University of Toronto. My research connects rigorous stochastic modelling with modern insurance data—from claim reserving and ratemaking to telematics and behavioural risk.
Professor · Department of Statistical Sciences
Associate Editor · IME, Annals of Actuarial Science & Risks
Research
From insurance portfolios to individual behaviour.
My work develops statistically principled methods that remain useful in practice—combining probability, Bayesian modelling, latent-state methods, machine learning, and granular insurance data.
Telematics & behavioural risk
High-frequency driving signals, hidden Markov models, anomaly detection, temporal motifs, collision detection, and dynamic risk indices.
Insurance analytics & ratemaking
Mixture-of-experts models, risk classification, random effects, censored and truncated loss data, and interpretable predictive modelling.
Mathematical risk theory
Dependence, ruin theory, Markov-additive processes, stochastic fluid flows, matrix-analytic methods, and operational risk.
Publications
Research since 2019.
Peer-reviewed and current research across telematics, reserving, ratemaking, insurance analytics, and stochastic risk modelling. Full-text links will be added as repository copies become available.
2026
A Portfolio-Anchored Frequency–Severity Risk Index for Trip and Driver Assessment Using Telematics Signals
Jongtaek Lee, Andrei Badescu & X. Sheldon Lin · ASTIN Bulletin
2026
Unified Bayesian Collective Reserving Framework with Hierarchical Temporal Smoothing
Hassan Abdelrahman, Andrei Badescu & X. Sheldon Lin · Insurance: Mathematics and Economics
2026
Marked Cox Models for IBNR Claims Count: Continuous and Discretized Approaches with Dirichlet-Driven Reporting Delays
Hassan Abdelrahman, Andrei L. Badescu, Radu V. Craiu & X. Sheldon Lin · ASTIN Bulletin 56(1): 60–88
2026
An Effective Bayesian GLM Approach for IBNR Claim Count Estimation
Hassan Abdelrahman, Andrei Badescu & X. Sheldon Lin · North American Actuarial Journal 30(3): 575–598
2025
Assessing Driving Risk Through Unsupervised Detection of Anomalies in Telematics Time Series Data
Ian Weng Chan, Andrei L. Badescu & X. Sheldon Lin · ASTIN Bulletin 55: 205–241
2025
Data Mining of Telematics Data: Unveiling the Hidden Patterns in Driving Behavior
Ian Weng Chan, Spark C. Tseung, Andrei L. Badescu & X. Sheldon Lin · North American Actuarial Journal 29(3): 275–309
2025
Claim Reserving via Inverse Probability Weighting: A Micro-Level Chain-Ladder Method
Sebastian Calcetero Vanegas, Andrei L. Badescu & X. Sheldon Lin · European Actuarial Journal 15: 89–135
2025
A Population Sampling Framework for Claim Reserving in General Insurance
Sebastian Calcetero Vanegas, Andrei L. Badescu & X. Sheldon Lin Submitted
2024
Effective Experience Rating for Large Insurance Portfolios via Surrogate Modeling
Sebastian Calcetero Vanegas, Andrei L. Badescu & X. Sheldon Lin · Insurance: Mathematics and Economics 118: 25–43
2023
Improving Risk Classification and Ratemaking Using Mixture-of-Experts Models with Random Effects
Spark C. Tseung, Ian Weng Chan, Tsz Chai Fung, Andrei L. Badescu & X. Sheldon Lin · Journal of Risk and Insurance 90: 789–820
2022
Fitting Censored and Truncated Regression Data Using Mixture-of-Experts Models
Tsz Chai Fung, Andrei L. Badescu & X. Sheldon Lin · North American Actuarial Journal 26(4): 496–520
2021
LRMoE.jl: A Software Package for Flexible Actuarial Loss Modelling Using Mixture-of-Experts Regression Models
Spark C. Tseung, Andrei L. Badescu, Tsz Chai Fung & X. Sheldon Lin · Annals of Actuarial Science 15(2): 419–440
2020
A New Class of Severity Regression Models with an Application to IBNR Prediction
Tsz Chai Fung, Andrei L. Badescu & X. Sheldon Lin · North American Actuarial Journal 25(2): 206–231
2019
A Class of Mixture-of-Experts Models for General Insurance: Theoretical Developments
Tsz Chai Fung, Andrei L. Badescu & X. Sheldon Lin · Insurance: Mathematics and Economics 89: 111–127
2019
A Class of Mixture-of-Experts Models for General Insurance: Application to Correlated Claim Frequencies
Tsz Chai Fung, Andrei L. Badescu & X. Sheldon Lin · ASTIN Bulletin 49(3): 647–688
2019
A Marked Cox Model for the Number of IBNR Claims: Estimation and Application
Andrei L. Badescu, Tianle Chen, X. Sheldon Lin & Dameng Tang · ASTIN Bulletin 49(3): 709–739
2019
Multivariate Cox Hidden Markov Models with an Application to Operational Risk
Tsz Chai Fung, Andrei L. Badescu & X. Sheldon Lin · Scandinavian Actuarial Journal 2019(8): 686–710
“Actuarial models should be mathematically coherent, statistically defensible, and useful for decisions.”
I joined the University of Toronto in 2006, following doctoral studies in actuarial science at Western University and postdoctoral work at the University of Waterloo. I became a full professor in 2020.
My earlier work focused on mathematical risk theory, queueing theory, stochastic fluid flows, and matrix-analytic methods. That foundation continues to shape my current research in claim reserving, ratemaking, insurance data science, and telematics.
I served as Director of the Master of Financial Insurance program from 2022 to 2025 and as Interim Chair of the Department of Statistical Sciences in 2024. I am an Associate Editor of Insurance: Mathematics and Economics (since 2018), the Annals of Actuarial Science (since 2024), and Risks (since 2026).
Collaboration & mentorship
Research through people.
My research program is built around long-term collaboration with students and colleagues, with projects designed to connect methodological contribution, reproducible computation, and actuarial relevance.
Insurance analytics
Flexible regression, mixture-of-experts modelling, latent heterogeneity, and software for modern loss data.
Reserving & pricing
Bayesian and sampling-based foundations for granular reserving, IBNR estimation, and coherent pricing.
Telematics risk
Signal-based detection of risky driving, temporal patterns, collision prediction, and interpretable behavioural indices.
Current & recent doctoral researchers
- Jongtaek LeeInsurance telematics and functional data
- Pak Hop ChanCyber-risk quantification and insurance applications
- Pat ChimtanooDoctoral research
- Yuxuan GuoDoctoral research
Selected graduates & former researchers
- Sophia Ian Wen ChanAssistant Professor, Nanyang University
- Hassan AbdelrahmanPostdoctoral Fellow, Department of Statistical Sciences, University of Toronto; incoming Assistant Professor, The American University in Cairo
- Sebastian Calcetero VanegasSenior Researcher, CIBC
- Tsz Chai FungAssistant Professor, Georgia State University
- Spark TseungLead Data Scientist, Reinsurance Group of America
MScAC industry supervision Applied research projects in partnership with Geotab, including collision severity modelling and automated load-weight detection for heavy-duty vehicles.
Current programs
Funded research with practical reach.
Current and recently completed projects connect statistical methodology with insurer and mobility applications, including telematics risk assessment, motif extraction, collision analysis, reserving, and ratemaking.
NSERC · 2026–2032
Integrating Statistical Modeling and Telematics Analytics for Advanced Insurance Risk Assessment
Individual Discovery Grant · Principal investigator
CAS · 2026–2027
Motif Extraction for Driving Behaviour Analysis and Insurance Ratemaking
Individual Research Grant · With Sophia Chan
CIA · 2026–2027
A Frequency–Severity Trip-Level Risk Index for Classification of Telematics Data
Academic Research Grant · With Jongtaek Lee
CIA · 2025–2026
Enhanced IBNR Reserve Estimation in P&C Insurance Through Bayesian Chain-Ladder Models
Academic Research Grant · With Hassan Abdelrahman
Knowledge translation Invited actuarial and industry presentations
Intellectual property Pending Canadian patent for telematics anomaly detection
Teaching
From foundations to contemporary insurance models.
My teaching spans undergraduate actuarial foundations and graduate work in risk modelling, credibility, simulation, and insurance analytics.
Graduate
Insurance Risk Models · Credibility and Simulation
Undergraduate
Mathematics of Finance · Mathematics of Investment and Credit · Life Contingencies · Loss Models · Advanced Topics in Actuarial Science
Supervision
Doctoral research and applied MScAC projects in reserving, ratemaking, telematics, stochastic modelling, and insurance data science, including industry-partnered projects with Geotab.
Contact
Research, supervision, and collaboration.
For academic enquiries, prospective research collaborations, or information about actuarial science at the University of Toronto, contact me at andrei.badescu@utoronto.ca.