Actuarial Science was born in the 17th century, and back then it was devoted to pricing contracts within life insurance. Latter, mostly in the 20th century, actuaries were able to get their methods applied to non-life insurance, too. Nowadays actuaries have been successfully evoking their skill-sets in the investment side of the insurance and banking industries. Thereby, modern Actuarial Science is a discipline that applies probability theory and statistics to assess and mitigate the risk of adverse financial events arising from financial market, catastrophic claims, pricing and reserving. York's actuarial mathematicians work on a variety of topics, covering pricing and modelling dependent risks, asset pricing and portfolio optimization, financial and insurance risk measurement and management. They have strong connections to colleagues in the Schulich School of Business, to research and industrial partners in the insurance/financial sectors as well as to such international professional actuarial organizations as the Casualty Actuarial Society and the Society of Actuaries.
John Tukey famously said that "the best thing about being a statistician is that you get to play in everyone's backyard." Our faculty's applied statistics research and collaborative endeavours embody this philosophy, spanning a wide range of fields such as medical imaging, biological data analysis, or finance. Our faculty collaborate with hospitals in Ontario, Quebec, and the U.S., and with Environment Canada. We also have past or ongoing collaborations with researchers in applied math, biology, the Centre for Disease Modelling, chemistry, computer science, and physics.
Our faculty specializes in the following sub-disciplines:
Statistics and Big Data
With continually decreasing costs of storing increasing amounts of data, massive data sets have become ubiquitous in today's world. "The [big data] revolution lies in improved statistical and computational methods, not in the exponential growth of storage or even computational capacity." Our faculty work on various methods including graphical models, composite likelihood for high-dimensional data, and machine learning algorithms. Applications include analysis of flu vaccine efficacy, West Nile virus and climate data, gene regulatory networks, and data fusion.
Biostatistics and Bioinformatics
Research in biostatistics and bioinformatics is grounded in the development and application of statistical methods to solve problems in biological and medical research. Our faculty work on a diverse set of issues including medical imaging, survival analysis, genomics, microarray and proteomic data analysis. The developed methods offer biologists and medical scientists new tools to analyze their data, and provide new insights and understanding of the information embedded in the biological data sets. Biostatistics and bioinformatics are fast evolving fields and which currently embrace the challenges and opportunities that come from Big Data. These fields not only solve practical science problems but also develop new statistical concepts and ideas.
Statistical Machine Learning
Machine learning is a vibrant field used in numerous fields such as finance, business, biology or medicine. Most of the methodology in machine learning is based on statistical ideas: statistics and machine learning are tightly interwoven. Many of the research topics at the intersection of statistics and machine learning are concerned with the transfer of statistical methods to the analysis of complex high-dimensional data. Statistical graphical models are one of the main tools in this area at the intersection of Statistics and computational sciences.
Statistical Methodology and Theory
We work on developing widely applicable novel methodologies while grounding these developments with mathematical foundations. Areas of interest are very broad, and include nonparametric and semiparametric methods, statistical inference and asymptotics, composite and empirical likelihood, clustering and classification, computational statistics, survival and longitudinal analysis, model selection, and graphical models for complex dependencies.