Home :: Clay Thompson


  • SIAM UQ 2018 I'll be attending the SIAM Conference on Uncertainty Quantification 2018. You can catch my brief review of methods for causal effect quantification on Thursday afternoon.  Abstract below.Researchers in applied statistics have developed a number of methods for the quantification of treatment effect size and variability from observational (i.e., non-experimental) data. However, the these methods suffer two major limitations. First, the proposed models may suffer from misspecifications as a result of marginalizing (unobserved latent confounders) and conditioning (non-random selection effects) relative to the data-generating model. Second, the data collected is often censored or incomplete. As a result of these uncertainties, inference methods associated with such models are inefficient at best and totally incorrect at worst ...
    Posted Apr 9, 2018, 6:45 AM by Clay Thompson
  • Academia vs Industry From Dan Simpson:  But I’m an academic statistician. And our key feature, as a people, is that we loudly and publicly dislike each other’s work. Even the stuff we agree with.  Why? Because people with our skills who also have impulse control tend to work for more money in the private sector.
    Posted Jan 2, 2018, 7:23 AM by Clay Thompson
  • Text highlighting during git commit message edit using vim When writing a git commit message, it is good practice to follow some rules regarding formatting. If you are a right-thinking person and use vim as your editor, this is made very easy--vim will highlight the text in different colors in accordance with these rules. Unfortunately, this was not working for me. After reading through the depths of the internet, the solution ended up being almost mindlessly simple. You just need to set the default editor (or, more generally, the editor that git is using) properly.
    Posted Nov 30, 2017, 5:22 PM by Clay Thompson
Showing posts 1 - 3 of 12. View more »

Basic Information

William Clayton Thompson ('Clay')

Senior Research Statistician Developer
Multivariable Models R&D
SAS Institute
Cary, NC, USA

Research Interests

Mathematical/computational modeling of physiological processes
Model calibration and validation for large-scale systems models
Modular coding, algorithm design, and data structures for large-scale models
Computational tools for reproducible research
Uncertainty quantification, variability modeling, rigorous model comparison
Mathematical and statistical aspects of inverse problems; statistical inference
Practical applications of functional analysis and measure theory
Scientific and mathematical methods to improve public policy

Recent Publications

See all Publications.

W. Clayton Thompson, Yingjiang Zhou, Saswata Talukdar, and C.J. Musante, PF-05231023, a long-acting FGF21 analogue, decreases body weight by reduction of food intake in non-human primates, JPKPD 43 (2016) 411-425. PMID 27405817

J. Selimkhanov, W.Clayton Thompson, T.S. Maurer, and C.J. Musante, Evaluation of a mathematical model of rat body weight regulation in application to caloric restriction and drug treatment studies, PLoS ONE 11 (2016) e0155674.

Recent Presentations

See all Presentations.