I was born in San Jose, California and grew up in Vancouver, Canada. I received a B.Sc. in Mathematics in 1986 and an M.Sc. in Statistics in 1989 from the University of British Columbia. In 1995, I received a Ph.D. in Statistics from the University of Chicago under the advising of Dr. Augustine Kong.
I joined the Department of Statistics at The Ohio State University in 1993, initially as an Instructor and then becoming an Assistant Professor in 1995. From 2000 to 2003, I was the Department's Statistical Computing Scientist. In September 2003, I moved to Cambridge and became a member of the Harvard Statistics Department as a Lecturer in Statistics.
Between December 2009 and September 2012, I was a Director of Marketing Science with in4mation insights, a market research, analytics and technology company based in Needham, MA. My focus with i4i was developing and implementing Bayesian solutions to market research problems. While with in4mation insights, major projects I worked on related to sensory science, retail store site location models, pricing models for internet retailers, and portfolio churn models for communications providers.
In September 2012, I became a Senior Statistician in the Data Sciences group at Compete, a market research company focusing on internet clickstream data located in Boston, MA. In July 2013, Compete joined with sister Kantar company Dynamic Logic to form MillwardBrown Digital (MBD). Compete's data comes from a statistically representative cross-section of 2 million consumers across the United States who have given permission to have their internet behaviors and opt-in survey responses analyzed anonymously as a source of marketing research. While with Compete, I led the development of MBD’s Consumer Behavioral Impact (CBI) and Path to Purchase (P2P) products. CBI efficiently estimates the lift of Internet advertising on Key Performance Indicators using a doubly robust propensity score based model to account for selection bias. P2P mines Internet clickstream data from a large multisource panel to cluster how people use the Internet to investigate products prior to purchase.
In 2017 I founded MEI Statistical Consulting, specializing in advanced research requiring customized research design, statistical analyses, and consultation.
My research interests in the past have included statistical genetics, focusing on problems in gene mapping, space-time hierarchical modeling, and statistical computing, dealing mainly with Monte Carlo methods. Past projects have dealt with modeling daily ozone data in the Lake Michigan region, class discovery and classification of tumor samples using gene expression data, the effects of interference in haplotype reconstruction, examining changes in military defensive strategies through point process models, and spatial-temporal nonlinear filtering in Command and Control.