Skip to content

Latest commit

 

History

History
69 lines (55 loc) · 4.25 KB

README.md

File metadata and controls

69 lines (55 loc) · 4.25 KB
Raw

Bayes2Business

Repository for publications including slides regarding Bayes in business/management

Goal

To introduce and enable Bayesian workflow in the business domain.

The rate of data inflow and decision outflow is increasingly getting faster and we need Bayesian workflow (BW). BW is defined as "three steps of model building, inference, and model checking/improvement, along with the comparison of different models, not just for the purpose of model choice or model averaging but more importantly to better understand these models." BW is needed for several reasons:

• Computation can be a challenge, and we often need to work through various steps including fitting simpler or alternative models, approximate computation that is less accurate but faster, and exploration of the fitting process, in order to get to inferences that we trust.

• In difficult problems we typically do not know ahead of time what model we want to fit, and even in those rare cases that an acceptable model has been chosen ahead of time, we will generally want to expand it as we gather more data or want to ask more detailed questions of the data we have.

• Even if our data were static, and we knew what model to fit, and we had no problems fitting it, we still would want to understand the fitted model and its relation to the data, and that understanding can often best be achieved by comparing inferences from a series of related models.

• Sometimes different models yield different conclusions, without one of them being clearly favourable. In such cases, presenting multiple models is helpful to illustrate the uncertainty in model choice.

This repository records the diffusion of BW in Business.

Resource

Events

AOM conference (08/01/2023)

We had two sessions on Bayes (PDW and Symposium) with the following content:

Professional Development Workshop

  1. Fundamental Advantages of Bayesian Statistics by David Krackhardt (Carnegie Mellon University)
  2. Executing Bayesian Mgt. Research by Mark Hansen (Brigham Young University)
  3. Bayesian Priors by Andreas Schwab (Iowa State University)
  4. Execution of Bayesian Analysis: MCMC and Software Solutions by Jeffrey Dotson (Brigham Young University) and Anup Nandialath (University of Wisconsin)
  5. Publishing Bayesian Studies in Management Journals by Mark Hansen (Brigham Young University)
  6. Persuading Reviewers and Editors to Publish Bayesian Analyses by Bill Starbuck (University of Oregon)

Symposium

  1. Advantages of Bayesian Statistics by David Krackhardt
  2. Illustrative Application of Bayesian Analyses by Andreas Schwab
  3. Why has Bayesian Analysis been used so little? by Bill Starbuck

With BW angle, Angie classified the nine based on why and how each presentation supports Bayesian Inference or Bayesian workflow in this repository. The abstract of each talk may be added.

gathering: bayes_aom

System dynamics seminar (10/20/2023)

List of supporters

(in order of joining)

  • Charlie Fine (MIT)
  • Jeffrey Dotson (Brigham Young Univ.)
  • Mark Hansen (Brigham Young Univ.)
  • David Krackhardt (Carnegie Mellon Univ.)
  • Andreas Schwab (Iowa State Univ.)
  • Bill Starbuck (Univ. of Oregon)
  • James Evans (Univ. of Chicago)
  • Andrew Gelman (Columbia Univ.)
  • Hazhir Rahmandad (MIT)
  • Anup Nandialath (Univ. of Wisconsin)
  • Özge Karanfil (Koç Univ.)
  • Matt Cronin (George Mason University)
  • Abdullah Almaatouq (MIT)

(invitation in progress)

  • Scott Stern (MIT)
  • Joshua Gans (Toronto Univ.)
  • Rahul Bhui (MIT)

Expectation from supporters

  • speaker of events to teach Bayes application and methodology (e.g. AOM or Bayesian entrepreneurship conference); Angie created a mailing list "bayes in business". Supporters would be automatically added to the mailing list, but feel free to unsubscribe.
  • (tbc)