Articles from:August 2026

  • Angela Ibhade Masters Thesis Defence: Wednesday, September 2, 1:00 PM

    Monday, August 31, 2026 | By

    Angela-Frances Osehiemen Ibhade, a Master of Science in Statistics candidate, will defend the thesis titled Markov-Switching Models: Applications in Credit Risk Analysis on Wednesday, September 2, 2026 from 1:00 PM– 3:30 PM in Plaza 600F.

    The examination committee includes Yifeng Li, Chair; Walid Ben Omrane (Goodman School of Business), External Examiner; William Marshall, Supervisor; Xiaojian Xu and Jan Vrbik, Supervisory Committee Members.

    Abstract:

    This thesis investigates the macroeconomic determinants of Canadian mortgage delinquency using a progression of time series models, culminating in a proposed mixed-lag Markov-switching vector autoregressive (MS-VAR) framework that allows autoregressive lag structure to vary across economic regimes. The analysis draws on 120 quarterly observations (1995 Q2-2024 Q4) covering mortgage delinquency, unemployment, GDP growth, inflation, and household debt servicing ratios.

    A vector error correction (VEC) model identifies four cointegrating relationships among the five variables, with significant error correction terms confirming that delinquency rates adjust toward a long-run macroeconomic equilibrium. A standard two-regime MS-VAR identifies a persistent baseline state and a transient stress state aligning with the 2008-2009 financial crisis and the COVID-19 pandemic of 2020, but near-zero stress-regime variance estimates reveal a degeneracy driven by the rarity of stress episodes in the sample. A proposed mixed-lag extension, allowing regime-specific lag orders via BIC grid search, selects an economically intuitive asymmetric structure but exhibits the same degeneracy, indicating the instability stems from limited stress-period data rather than lag misspecification alone. The VEC framework is therefore recommended as the preferred specification, producing interpretable dynamics and stable forecasts over a 50-quarter horizon. Both the standard and mixed-lag MS-VAR frameworks are implemented in the MSMVAR R package,1 publicly released as open-source software on GitHub, providing a foundation for future applications with larger samples or clearer regime separation.

    Keywords: Markov-switching VAR, mixed-lag models, mortgage delinquency, cointegration, EM algorithm, Canadian credit risk, regime-switching, BIC model selection, open-source software

  • Ryan Gunther Masters Thesis Defence: Wednesday, August 12, 11:00 AM

    Tuesday, August 04, 2026 | By

    Ryan Gunther, a Master of Science in Statistics candidate, will defend his thesis titled Dimensionality Reduction Approaches for Baseball Swing Motion Analysis on Wednesday, August 12, 2026 at 11:00 AM in GSB 405.

    The examination committee includes Melanie Pilkington, Chair; Shawn Beaudette, External Examiner; William Marshall, Supervisor; Jan Vrbik and S. Ejaz Ahmed, Supervisory Committee Members.

    Abstract:

    Background: Baseball biomechanical data are often high-dimensional, temporally structured and difficult to analyze directly. Although baseball biomechanics research has been heavily invested into over the past 10 years, the majority of it has been to the benefit of pitchers rather than batters. This thesis aims to investigate whether dimensionality reduction methods for large baseball biomechanics datasets of batters could successfully create compact representations of swing motion while preserving information relevant to bat speed. 

    Methods: First, dimensionality reduction techniques (principal-feature selection, probabilistic principal component analysis, covariance-based temporal segmentation, and clustering) were performed on a public biomechanics dataset of 431 marker-based motion capture swings, reducing them to feature vectors to perform cross-validated LASSO regression on. Separately, a computer vision pipeline and corresponding 200 swing MLB dataset were built. Hitter silhouettes were segmented, transformed into distance-weighted images, and embedded as two-dimensional Isomap trajectories, again with the goal of extracting meaningful features from a reduced dataset. Ridge regression and recurrent neural networks were trained on the data (simple RNN, GRU, and LSTM) to examine predictive performance.

    Results: Both datasets were successfully reduced to a compact representation that could be used for visualization, comparison, and downstream modelling, but predictive gains were limited. Across the marker-based and markerless datasets, models produced little improvement over the mean baselines. Adding first- and second- derivative features and adjusting bat speed for pitch context did not materially change this result. 

    Conclusion: The representations created a bridge between raw motion data and interpretable analysis, allowing swing movement to be studied in forms that would be difficult to access from the original high-dimensional inputs alone. In the available datasets, the reduced representations did not support strong out-of-sample prediction of bat speed (or exit velocity), but they provide a framework for organizing complex motion data for future analysis.