Calvert Jump, Robert ORCID: https://orcid.org/0000-0002-2967-512X (2024) Tractable Bayesian inference for an unidentified simple linear regression model. The American Statistician. ISSN 0003-1305 (Print), 1537-2731 (Online) (doi:10.1080/00031305.2024.2333864)

Preview	PDF (Open Access Article) 46413_CALVERT JUMP_Tractable_Bayesian_inference_for_an_unidentified_simple_linear_regression_model_(OA)_2024.pdf - Published Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (1MB) | Preview
Preview	PDF (AAM) 46413_CALVERT JUMP_Tractable_Bayesian_inference_for_an_unidentified_simple_linear_regression_model.pdf - Accepted Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (536kB) | Preview

Abstract

In this paper, 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 t-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 t-distribution. This result is closely related to the literatures on unidentified regression models, imperfect instrumental variables, and sensitivity analysis.

Item Type:	Article
Uncontrolled Keywords:	identification; unidentified models; sensitivity analysis; Bayesian statistics; omitted variable bias; linear regression
Subjects:	H Social Sciences > H Social Sciences (General) H Social Sciences > HA Statistics Q Science > QA Mathematics
Faculty / School / Research Centre / Research Group:	Faculty of Business Greenwich Business School > Political Economy, Governance, Finance and Accountability (PEGFA)
Last Modified:	02 Dec 2024 16:08
URI:	http://gala.gre.ac.uk/id/eprint/46413

Actions (login required)

View Item

Downloads