Larry A. Wasserman

To me, what's interesting about the field of statistics is the interplay between applications, methodology and theory. My particular research interests include model selection, nonparametric inference, causal inference, Bayesian methods, and biostatistics.

An underlying theme in all my work is the notion of robustness. A statistical method is robust if it is not sensitive to underlying assumptions. Model selection techniques achieve this goal by allowing us to consider a range of possible models for a set of data instead of a single model. Nonparametric methods go a step further by replacing traditional finite dimensional models with more flexible, infinite dimensional models. Causal inference provides a set of tools for exposing subtle assumptions that are at play in numerous inference problems. For example, to estimate the effect of a treatment on survival we often need to account for the role of confounding variables. Causal inference provides the basis for understanding the effect these confounders.

In all these areas, I am guided by applications in biostatistics and other fields. Typically, statistical computing is a crucial element in most of these applications. For example, I am working with cognitive psychologists on a research project in which we are using modern Bayesian methods to compare complicated models that have been proposed to explain human cognition. Recent statistical computing methods play an important role in this project. Indeed, the methods would have been impossible to implement even a few years ago.

Some Related Publications

Robins, J. and Wasserman, L. (1996) On the Impossibility of Inferring Causation From Association Without Background Knowledge. To appear.

Barron, A., Schervish, M. and Wasserman, L. (1996) The Consistency of Posterior Distributions in Nonparametric Problems. Technical report.

Kass, R.E. and Wasserman, L. (1996) "The selection of prior distributions by formal rules," Journal of the American Statistical Association, 91, pp. 1343-1370.

Kadane, J.B. and Wasserman, L. (1996) "Symmetric, Coherent, Choquet Capacities," Annals of Statistics, 24, pp. 1250-1264.