# Welcome to the Department of Statistics and Actuarial Science

The Department of Statistics and Actuarial Science is a top tier academic unit among statistical and actuarial science globally. Our students and faculty explore topics such as Actuarial Science, Biostatistics, Data Science, Quantitative Finance, Statistics, & Statistics-Computing. Our department is home to:

50+

full-time faculty researching diverse and exciting areas

1000+

undergraduate students from around the world

175+

graduate students in Master, Doctoral, and professional programs

Interested in graduate studies with Statistics and Actuarial Science? Meet Mingyu (Bruce) Feng, a PhD student in actuarial science. Bruce is a pioneer researcher in the field of sustainable investment and the impact of climate change. Learn more about furthering your education on the Future Graduate page on the Math site.

1. Oct. 18, 2019Maclean’s features READI

Maclean’s wrote an article about how the University of Waterloo and funding from the Canadian government is helping Indonesia prepare for climate change. Through the Risk Management, Economic Sustainability and Actuarial Science in Indonesia (READI)project in the Department of Statistics and Actuarial Science, Waterloo is working to establish Indonesia as a regional centre of actuarial science.

Indonesia’s universities did not have a top-ranked actuarial science program, nor a cooperative program. With funding from Global Affairs, Manulife Indonesia, and Sun Life, the dedication of staff and faculty members from the University of Waterloo, and eager partners in Indonesia, nine out of 12 Indonesian universities are working with READI.

The success continues to grow. At the most recent co-op event, more than 80 companies showed up at one university. To learn more about the success of the program, see the full article in Maclean’s

2. Oct. 11, 2019Interested in Actuarial Science?

Learn about the program from Angel Yang, a fourth year student majoring in Actuarial Science at the University of Waterloo. In her interview with AdvisorSmith, Angel talks about why she chose to study at the University of Waterloo, her experience on campus, why a co-op program made sense to her, and much more. View the full interview online to see what she has to say.

If you are interested in pursuing Actuarial Science, Waterloo should be at the top of your list. The Department of Statistics and Actuarial Science (SAS) is considered a top tier academic unit in the field of Actuarial Science.  Our department has one of the world’s largest programs at both the undergraduate and the graduate levels. SAS offers professional masters programs as well as research-oriented masters and doctoral programs in actuarial science and finance. It is home to the bachelor of mathematics in actuarial science program which covers a wide range of courses, including full coverage of the material of the SOA/CAS associateship requirements and some coverage of the SOA/CAS fellowship requirements. With a sizeable actuarial faculty, the range of courses offered is broad and extends well beyond the SOA/CAS syllabi. Students may choose to gain foundational knowledge in life insurance, property and casualty insurance, pensions, risk theory, quantitative finance, and corporate finance. Students who are particularly interested in financial or predictive analytics topics may elect to add either option to their actuarial science honours plan.

3. Sep. 20, 2019Celebrating Mary Thompson's 50th anniversary in the Department of Statistics and Actuarial Science

On Wednesday, September 18, 2019, the Department of Statistics and Actuarial Science (SAS) celebrated the 50th anniversary of Professor Mary Thompson as a faculty member in the department. Many current faculty and staff and a number of retired professors gathered in the SAS Lounge and celebrated Mary’s milestone with cake, coffee and fruits. The Vice-President, Research & International, of University of Waterloo, Professor Charmaine Dean, who was a PhD graduate from the department, sent a beautiful bouquet to congratulate Mary on this special occasion. A couple retired faculty members turned the clock back and told stories about Mary during the early days of the department. Mary thanked the department for being her academic home for the past 50 years, and told the younger generation of faculty members that SAS is a “can-do” place to fulfill their full potential and aspiration.

Mary joined the department in 1969 as one of the first group of statistics faculty members, when the department was still in its infancy (established in 1967). Over the past half a century, Mary has become a fixture of the department, a highly accomplished scholar, and a great inspiration and role model for many students and young faculty members. She has provided dedicated services to the statistical community at all levels, including chair of the department, acting dean of the faculty, first scientific director of the Canadian Statistical Sciences Institute, and president of the Statistical Society of Canada.

1. Oct. 25, 2019Department seminar by Fabio Bellini, Università degli Studi di Milano-Bicocca

On the properties of Lambda-quantiles

We present a systematic treatment of Lambda-quantiles, a family of generalized quantiles introduced in Frittelli et al. (2014) under the name of Lambda Value at Risk. We consider various possible definitions and derive their fundamental properties, mainly working under the assumption that the threshold function Lambda is nonincreasing. We refine some of the weak continuity results derived in Burzoni et al. (2017), showing that the weak continuity properties of Lambda-quantiles are essentially similar to those of the usual quantiles. Further, we provide an axiomatic foundation for Lambda-quantiles based on a locality property that generalizes a similar axiomatization of the usual quantiles based on the ordinal covariance property given in Chambers (2009). We study scoring functions consistent with Lambda-quantiles and as an extension of the usual quantile regression we introduce Lambda-quantile regression, of which we provide two financial applications.

(joint work with Ilaria Peri).

2. Oct. 31, 2019Department seminar by Qi Long, University of Pennsylvania

Variable selection for structured high-dimensional data using known and novel graph information

Variable selection for structured high-dimensional covariates lying on an underlying graph has drawn considerable interest. However, most of the existing methods may not be scalable to high dimensional settings involving tens of thousands of variables lying on known pathways such as the case in genomics studies, and they assume that the graph information is fully known. This talk will focus on addressing these two challenges. In the first part, I will present an adaptive Bayesian shrinkage approach which incorporates known graph information through shrinkage parameters and is scalable to high dimensional settings (e.g., p~100,000 or millions). We also establish theoretical properties of the proposed approach for fixed and diverging p. In the second part, I will tackle the issue that graph information is not fully known. For example, the role of miRNAs in regulating gene expression is not well-understood and the miRNA regulatory network is often not validated. We propose an approach that treats unknown graph information as missing data (i.e. missing edges), introduce the idea of imputing the unknown graph information, and define the imputed information as the novel graph information.  In addition, we propose a hierarchical group penalty to encourage sparsity at both the pathway level and the within-pathway level, which, combined with the imputation step, allows for incorporation of known and novel graph information. The methods are assessed via simulation studies and are applied to analyses of cancer data.

3. Nov. 7, 2019Department seminar by Yang Ning, Cornell University

Nonregular and Minimax Estimation of Individualized Thresholds in High Dimension with Binary Responses

Given a large number of covariates $\bZ$, we consider the estimation of a high-dimensional parameter $\btheta$ in an individualized linear threshold $\btheta^T\bZ$ for a continuous variable $X$, which minimizes the disagreement between $\sign{X-\btheta^T\bZ}$ and a binary response $Y$. While the problem can be formulated into the M-estimation framework, minimizing the corresponding empirical risk function is computationally intractable due to discontinuity of the sign function. Moreover, estimating $\btheta$ even in the fixed-dimensional setting is known as a nonregular problem leading to nonstandard asymptotic theory. To tackle the computational and theoretical challenges in the estimation of the high-dimensional parameter $\btheta$, we propose an empirical risk minimization approach based on a regularized smoothed non-convex loss function. The Fisher consistency of the proposed method is guaranteed as the bandwidth of the smoothed loss is shrunk to 0. Statistically, we show that the finite sample error bound for estimating $\btheta$ in $\ell_2$ norm is $(s\log d/n)^{\beta/(2\beta+1)}$, where $d$ is the dimension of $\btheta$, $s$ is the sparsity level, $n$ is the sample size and $\beta$ is the smoothness of the conditional density of $X$ given the response $Y$ and the covariates $\bZ$. The convergence rate is nonstandard and slower than that in the classical Lasso problems. Furthermore, we prove that the resulting estimator is minimax rate optimal up to a logarithmic factor. The Lepski's method is developed to achieve the adaption to the unknown sparsity $s$ and smoothness $\beta$. Computationally, an efficient path-following algorithm is proposed to compute the solution path. We show that this algorithm achieves geometric rate of convergence for computing the whole path. Finally, we evaluate the finite sample performance of the proposed estimator in simulation studies and a real data analysis from the ChAMP (Chondral Lesions And Meniscus Procedures) Trial.

All upcoming events

## Cyntha Struthers

Associate Professor

Math Faculty Teaching Fellow

Contact Information:
Cyntha Struthers

Research interests

Methods for modelling continuous-time longitudinal data subject to missingness.