I am a Quantitative Psychology post-doc at University of California, Davis. I enjoy building new models and extending old ones to improve predictions, decision making, and inferences.
Upon discovering and immersing myself in Bayesian statistics, my research focus substantially shifted from social psychology toward building better models and making more powerful inferences. In teaching myself how to do Bayesian GLMs, GLMMs, latent variable models, mixture models, location-scale-shape models, meta-analysis, integrated data analysis, and many others, I realized that the statistical world of social psychology could be significantly improved (no pun intended). With fancier theoretical models, freer assumptions, and a potent probabilistic framework, researchers can gain more information (substantively), more information (in the Fisher sense of the word), make better predictions and decisions, and generally create and compare models with as many or as few assumptions, with as little or as much complexity as desired. Of no surprise, Stan is my tool of choice for building and testing new models for inference and probabilistic decision making.
My interests include Bayesian SEM (with normal or non-normal assumptions), robust modeling (principled handling of skew, “outliers”, heteroskedasticity), measurement models and invariance assessment, improving analytic techniques for common methods (e.g., the SC-IAT), and in general assessing performance of novel modeling techniques/assumptions and inferential methods through simulation. We are currently using the mixed effects location scale model (MELSM) to explore new research venues. This includes better measurement models, including intraindividual variability as an outcome and predictor in SEMs, and much more.
Feel free to contact me for consulting, subcontracting, or positions.