Professor Wing Hung Wong serves on the faculty of Stanford University where he is currently Professor of Statistics, Professor of Biomedical Data Science, and Stephen R. Pierce Family Goldman Sachs Professorship in Science & Human Health. Before joining Stanford in 2004, he held teaching positions at the University of Chicago, The Chinese University of Hong Kong, UCLA and Harvard University. He chaired the Stanford Department of Statistics from 2009 to 2012. Professor Wong was the winner of the COPSS Presidents’ Award (1993) and the COPSS Distinguished Achievement Award (2021). He was elected to the US National Academy of Sciences in 2009 and the Academia Sinica in 2010. He was a founding member of the Hong Kong Academy of Sciences in 2015.
His past research contributions include:
1. mathematical statistics, where he clarified the large sample properties, in general parameter spaces, of likelihood functions and of sieve maximum likelihood estimates;
2. Bayesian statistics, where he introduced sampling-based algorithms into Bayesian computational inference; and
3. computational biology, where he developed tools for the analysis of microarrays and sequencing data and applied them to study gene regulatory systems.
Technologies from his group have led to the formation of several companies in the space of genomics data analysis and genomic medicine.
His current research interests include the use of deep neural networks in mainstream statistical areas such as density estimation, clustering, Bayesian analysis and causal inference; and the development of novel genomics technologies based on semiconductors. In the latter, his group recently created a platform for 2D control of droplets that will enable complex cell biology experiments to be performed independently but in a massively parallel and programmable manner. He believes that the design and analysis of such experiments will present new opportunities for statistical research.
This will be the second IMS Grace Wahba Lecture (the first was given at the IMS annual meeting in London last year by Michael Jordan). It will be given at the Joint Statistical Meetings in Toronto, August 5–10, 2023.

Causal inference by encoding generative modeling

We consider the problem of inferring the causal effect of an exposure variable X on an outcome variable Y. Besides X and Y, a high-dimensional covariate V is also measured. It is assumed that confounding variables that may cause bias in the desired causal inference are low-dimensional features of V. Under this assumption, we propose an encoding generative modeling (EGM) approach for the estimation of the average dose response function, a function that captures, in an average sense, the dependency of Y on X when confounders were hold fixed.

We show that EGM provides a framework for us to develop deep learning-based estimates for the structural equations that describe the causal relations among the variables. We will present numerical and theoretical evidence to demonstrate the effectiveness of our approach.