Choose Language Hide Translation Bar
Peter Lenk shares his journey from math, to statistics, to Bayesian and beyond

PeterLEnk.pngProfessor Peter Lenk of the University of Michigan is an expert on Bayesian methods.Peter Lenk has a richly storied journey — from influences in his childhood that led him to pursue a PhD in statistics to teaching and groundbreaking research for useful applications in business and economics. Peter is Professor of Technology and Operations and Marketing at the Stephen M. Ross School of Business at the University of Michigan. He has made many contributions to Bayesian inferential, forecasting, and semiparametric modeling methods.

His teaching spans management decision analysis, predictive modeling and forecasting. He is also a fellow of the American Statistical Association, and has many other honors and awards.

Peter is our guest on this month’s episode of Analytically Speaking. He shares what attracted him to Bayesian statistics — in part because it is prescriptive about what should be done in any situation.  

We use JMP Text Explorer to visualize some of his publications, and it's no surprise that Bayesian terms are prominent. Though he has many publications and citations, the outlier cited the most is Hierarchical Bayes Conjoint Analysis: Recovery of Partworth Heterogeneity from Reduced Experimental ..., co-authored with DeSarbo et al. He also shares the serendipitous circumstances that led him to apply Bayesian methods to marketing applications.

His first encounter with JMP (one of the very early releases) didn’t win him over. But a decade or so later, the University of Michigan adopted JMP, and Peter has given us a wonderful customer story on what he likes about teaching with JMP. (He tells his students it’s a treasure hunt to see what’s behind the red triangles.)

You can learn more about what Peter has to share by watching this month’s episode of Analytically Speaking, and check out the customer story, Sophisticated analytics made easy.

 

Screen Shot 2018-06-08 at 1.19.01 PM.png

Article Labels