Welcome to the Department of Statistics and Actuarial Science

The Department of Statistics and Actuarial Science is among the top academic units for statistical and actuarial science in the world and is home to more than 40 research active full-time faculty working in diverse and exciting areas. The Department is also home to over 900 undergraduate students and about 150 graduate students in programs including Actuarial Science, Biostatistics, Quantitative Finance, Statistics, and Statistics-Computing.

We are located on University of Waterloo main campus, which is located at the heart of Canada's Technology Triangle about 100 kilometers west of Toronto.

  1. July 16, 2018Nursing notes can help indicate whether ICU patients will surviveNurse writing up her notes

    Researchers at the University of Waterloo have found that sentiments in the nursing notes of health care providers are good indicators of whether intensive care unit (ICU) patients will survive. 

    Hospitals typically use severity of illness scores to predict the 30-day survival of ICU patients. These scores include lab results, vital signs, and physiological and demographic characteristics gathered within 24 hours of admission. 

  2. July 16, 2018Congratulations Professor Ruodu Wang, recipient of the NSERC DAC award!Ruodu Wang

    Professor Ruodu Wang received a $120,000 Discovery Accelerator Supplement (DAC) from the Natural Sciences and Engineering Research Council (NSERC) for a proposal titled “Model Uncertainty and Robustness in Risk Management.”

  3. July 10, 2018The Department of Statistics and Actuarial Science is pleased to welcome Assistant Professor Samuel Wong as of July 1st 2018.Samuel Wong

    Samuel Wong (PhD 2013, Harvard University) joins our department from the University of Florida where he was an Assistant Professor.  Samuel's research focuses on developing analytical methods to tackle data-driven problems arising in scientific domains. Currently, his main applications of interest are protein structure prediction, learning dynamic systems in biology, and quality assessment of forest products. Statistical areas featured in his work include Bayesian modeling, Monte Carlo methods, and approximate inference strategies. With his data science focus, Samuel is keen on solving problems arising through collaboration, where both principled methodology and large-scale computation are needed.  To learn more about Samuel's research, please visit his website.

Read all news
  1. July 19, 2018Department seminar by Geneviève Gauthier, HEC Montreal

    Extracting Latent States from High Frequency Option Prices

    We propose the realized option variance as a new observable variable to integrate high frequency option prices in the inference of option pricing models. Using simulation and empirical studies, this paper documents the incremental information offered by this realized measure. Our empirical results show that the information contained in the realized option variance improves the inference of model variables such as the instantaneous variance and variance jumps of the S&P 500 index. Parameter estimates indicate that the risk premium breakdown between jump and diffusive risks is affected by the omission of this information.

  2. Aug. 2, 2018Department seminar by Daniel Farewell, Cardiff University

    No Such Thing as Missing Data

    The phrase "missing data" has come to mean "information we really, really wish we had". But is it actually data, and is it actually missing? I will discuss the practical implications of taking a different philosophical perspective, and demonstrate the use of a simple model for informative observation in longitudinal studies that does not require any notion of missing data.

  3. Aug. 8, 2018Department seminar by Yang Li, Renmin University

    Model Confidence Bounds for Variable Selection

    In this article, we introduce the concept of model confidence bounds (MCBs) for variable selection in the context of nested models. Similarly to the endpoints in the familiar confidence interval for parameter estimation, the MCBs identify two nested models (upper and lower confidence bound models) containing the true model at a given level of confidence. Instead of trusting a single selected model obtained from a given model selection method, the MCBs proposes a group of nested models as candidates and the MCBs’ width and composition enable the practitioner to assess the overall model selection uncertainty. A new graphical tool — the model uncertainty curve (MUC) — is introduced to visualize the variability of model selection and to compare different model selection procedures. The MCBs methodology is implemented by a fast bootstrap algorithm that is shown to yield the correct asymptotic coverage under rather general conditions. Our Monte Carlo simulations and a real data example confirm the validity and illustrate the advantages of the proposed method.

All upcoming events

Faculty Joint Publications

Map of Faculty and PhD Students backgrounds

Meet our people

Riley Metzger

Riley Metzger


Contact Information:
Riley Metzger

Research interests

Statistics education.