You are here:


The theoretical alignment of Bayesian statistics and cumulative prospect theory

The focus of this research is combining Bayesian statistics, in particular Bayesian data analysis, with utility theory. From utility theory we select the work from Kahneman and Tversky, i.e., Cumulative Prospect Theory. Computer science and software engineering research today focuses on traditional frequentist approaches from statistics, to point at improvements a certain technique, approach, method, has compared to a baseline. We believe that this approach is used too often to represent a dualism, i.e, ending up in a binary decision: yes/no, 0/1, or pass/fail. We argue that reality is rarely this simple. Instead, by combining Bayesian data analysis with cumulative prospect theory, we believe we can o er realistic scenarios to decision makers, which allows them to receive a better understanding of where the borders between yes/no are and how they a ffect their processes, organizations, etc. To this end, many times they might not select a `0′ or `1′, but rather pick `0.5′.


Fellows involved in this project


Related publications

Journal Article

Torkar, Richard, Carlo Alberto Furia, Robert Feldt, Francisco Gomes de Oliveira Neto, Lucas Gren, Per Lenberg and Neil A. Ernst. 2021. A Method to Assess and Argue for Practical Significance in Software Engineering. IEEE Transactions on Software Engineering, 1–1.

Journal Article

Ali, Nauman bin, Henry Edison and Richard Torkar. 2020. The Impact of a Proposal for Innovation Measurement in the Software Industry. ACM / IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM).

Share this project:

Share on whatsapp
Share on email
Share on facebook
Share on twitter
Share on linkedin

Is any information on this page incorrect or outdated? Please notify Ms. Nel-Mari Loock at [email protected].