.png){kind=icon}
.png){kind=icon}
Return
Empirical Bayes When Estimation Precision Predicts Parameters
.png){kind=icon}
PRE
AI
J
DOI：10.3982/ECTA22935.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Gaussian empirical Bayes methods usually maintain a precision independence assumption: The unknown parameters of interest are independent from the known standard errors of the estimates. This assumption is often theoretically questionable and empirically rejected. This paper proposes to model the conditional distribution of the parameter given the standard errors as a flexibly parameterized location-scale family of distributions, leading to a family of methods that we call close. The close framework unifies and generalizes several proposals under precision dependence. We argue that the most flexible member of the close family is a minimalist and computationally efficient default for accounting for precision dependence. We analyze this method and show that it is competitive in terms of the regret of subsequent decision rules. Empirically, using close leads to sizable gains for selecting high-mobility Census tracts.
Keywords:
Empirical Bayes
g-modeling
regret
heteroscedasticity
nonparametric maximum likelihood
Opportunity Atlas
Creating Moves to Opportunity
Journal
IF:
7.1
Papers:
3.0K
Citations:
4.3W
Organization
Cited Papers