Lam Ho



Canada Research Chair in Stochastic Modelling

Tier 2 - 2017-11-01
Dalhousie University
Social Sciences and Humanities Research Council

902-494-2093
lam.ho@dal.ca

Coming to Canada From


University of California, Los Angeles, USA

Research involves


Developing much-needed theory, and building statistical software for partially observed stochastic models.

Research relevance


This research will lead to the development of novel statistical methods which will advance the study of evolutionary biology and epidemiology through stochastic modeling.

Managing the vast space of unobservable events


Real-world problems are usually very complicated because there are more events that are unobservable than observable. For example, biologists want to learn about the evolutionary history of life but can only observe present-day and fossil organisms. Similarly, epidemiologists often study the spread of infectious pathogens using daily, weekly, monthly or yearly incidence counts since it is very unlikely that they can know exactly when an individual was infected.

When the space of unobservable events is huge, taking into account this uncertainty is extremely difficult. Existing theories and methods are not capable of handling problems with massive amounts of unobservable events. These scenarios require new approaches.

Dr. Lam Ho, Canada Research Chair in Stochastic Modelling, is developing novel statistical theory and methods for partially observed stochastic (probability distribution) models. In particular, he is investigating the statistical properties of methods for cross-species analysis, and establishing methods for weeding out the unobservable events of infectious disease epidemics. Ho's lab is also building high-performance software for data analysis based on new findings.

This research will provide much-needed tools for scientists to tackle problems where the space of unobservable events is large. Specifically, the results will directly help researchers in studying evolutionary history and combating infectious diseases through stochastic models.