Back to Search
Start Over
Tractable Bayesian Inference For An Unidentified Simple Linear Regression Model.
- Source :
-
American Statistician . Mar2024, p1-12. 12p. 2 Illustrations. - Publication Year :
- 2024
-
Abstract
- AbstractIn this article, I propose a tractable approach to Bayesian inference in a simple linear regression model for which the standard exogeneity assumption does not hold. By specifying a beta prior for the squared correlation between an error term and regressor, I demonstrate that the implied prior for a bias parameter is <italic>t</italic>-distributed. If the posterior distribution for the identified regression coefficient is normal, this implies that the posterior distribution for the unidentified treatment effect is the convolution of a normal distribution and a <italic>t</italic>-distribution. This result is closely related to the literatures on unidentified regression models, imperfect instrumental variables, and sensitivity analysis. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 00031305
- Database :
- Academic Search Index
- Journal :
- American Statistician
- Publication Type :
- Academic Journal
- Accession number :
- 176241669
- Full Text :
- https://doi.org/10.1080/00031305.2024.2333864