Interim Analysis and Stopping Rules

In some experimental studies, because of issues like cost or safety, it may be advantageous to stop data collection prior to the planned study end. In Bayesian statistics, thanks to the stopping rule principle, inferences are not changed by the choice of when to end a study. Interim analysis methods let us evaluate a study's expected probability of success/futility based on pre-specified decision rules while the study is ongoing, and can allow for early stopping of studies when their eventual results seem clear prior to the planned study end.

Incorporating Informative Priors

Another way to improve research efficiency comes from the inclusion of informative priors in Bayesian analyses. My work with informative priors focuses on two approaches: eliciting expert information from subject matter experts using conditional means priors, and incorporating prior study data using power priors.

Model Selection

George Box famously said, "All models are wrong, but some are useful." Model selection determines which among a collection of candidate models are most useful for a particular scientific task (e.g. hypothesis generation, prediction). Research on Bayesian model selection takes many forms, from the analysis of Bayesian model selection criteria to the careful use of informative prior distributions to learn more about posterior probabilities for various models.

Decision Theory

Decision theory is the one-player version of game theory. In a given situation, it relates the actions one can take to the unknown states of nature that will determine the consequences of taking that action. The discipline of decision theory attempts to choose the optimal action under uncertainty, given one's valuation of the consequences that may ensue and one's beliefs about the probability that those consequences will eventuate. Decision theory underlies a great deal of the development of modern statistics.

Philosophical Foundations

Good statistical thinking depends on our understanding of epistemology and the philosophical roots of scientific inquiry. How do we know what we know? What constitutes knowledge? Bayesian notions of probability and learning provide (in my opinion) the best solution to Hume's problem of induction. Statistics functions best when informed by a sense of how statistical methods address important difficulties in the rational acquisition of knowledge.

Mind Research Network

The Mind Research Network (MRN) is an independent non-profit organization that focuses on imaging technology for neuroscience investigation. I am currently working with them on an NIH R33 grant looking at the effectiveness of music therapy on children and adolescents with fetal alcohol syndrome disorder. My previous work with MRN has involved assessing functional connectivity differences between healthy controls and patients with schizophrenia.

Sandia National Laboratories

My work with Sandia National Labs has focused on anomaly detection methods—especially in cost-constrained situations. This work has variously involved feature selection, multi-attribute utility theory, multi-channel time-series data, and applications in cybersecurity.

US Air Force

With my former student (Maj Victoria R.C. Sieck, PhD) I have done work on interim analysis and adaptive design methods for operational testing scenarios.

Other Collaborations

I have done some other collaborations with academic partners at UC Irvine and UNM, especially in applications involving population health and neuroimaging. I am always interested in collaborating on interesting problems with Bayesian angles to them, especially in biostatistical and psychometric domains where I have a bit more knowledge going in. If you have a problem you think I might find interesting, please let me know by email.

Wikihistory

by Desmond Warzel

Extracts from the International Association of Time Travelers: Members' Forum.

A Tall Tail

by Charles Stross

Interesting stories get told at scientific conferences.

Firstborn

by Brandon Sanderson

The pain of living in your brother's shadow, except in space with lasers.

The Last Question

by Issac Asimov

"How can the net amount of entropy of the universe be massively decreased?"

Writing Excuses Podcast

This podcast, now in its 12th season, is run by four high-profile genre fiction authors: Brandon Sanderson, Dan Wells, Howard Tayler, and Mary Robinette-Kowal. Not only are their insights pretty phenomenal, but they regularly bring in guests for alternative perspectives. Podcast episodes run about 15 minutes—but they always end with a writing prompt, so you might wind up spending more time than that on each episode.

Brandon Sanderson's BYU Writing Class

Brandon Sanderson, one of the hosts of the Writing Excuse podcast and my favorite author, teaches a yearly creative writing class at Brigham Young University (BYU). His lectures occasionally get recorded and posted online (with his endorsement).