Layman’s abstract for paper on a Bayesian approach to regression modelling with the tilted beta distribution

Each week, we will be publishing layman’s abstracts of new articles from our prestigious portfolio of journals in statistics. The aim is to highlight the latest research to a broader audience in an accessible format.

The article featured today is from the Canadian Journal of Statistics, with the full article now available to read on Early View here.

Hahn, E.D. (2020), Regression modelling with the tilted beta distribution: A Bayesian approach. Can J Statistics. doi:10.1002/cjs.11563

Test grades, SAT scores, government percentages, and many other types of data range from a minimum to a maximum. Often this range is from zero to 100% such as in the case of test scores. Surprisingly, the commonly used statistical regression models for this data treat the maximum and the minimum entirely differently from the rest of the data. For example, if you received a 99.999% on a test, the commonly used models would treat you entirely differently from a person who got a 100%, and vice versa. In this paper the author introduces a new regression model where the maximum and minimum are treated as part of the entire data continuum. This appears to be a new advance in statistical modeling. The author finds that Bayesian methods are especially useful for this type of model.