Skip to content

Bias reduced fixed effect generalized linear models

License

Notifications You must be signed in to change notification settings

JohannesSKunz/brfeglm

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

31 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

BRFEGLM: estimate bias reduced fixed effect glm in Stata.


Overview

brfeglm is a Stata command that estimates bias-reduced fixed effects glm models for probit, logit and cloglo by iterative weighted least squares (IWLS) and with a large dummy-variable set.

The program builds upon brglm, but is tailored to the estimation of fixed effects, it is much faster, omits the fixed effects from the regression output but extracts and stores them automatically in a new variables.

Works with margin command.

In combination with the package fese_fast it further allows to also get an estimate of the fixed effects standard error.

Installation

Can be installed in STATA via:

* Install the most recent version of -brfeglm-
net install brfeglm, from("https://raw.githubusercontent.com/JohannesSKunz/brfeglm/master") replace

Example

Here is the Stata script:

webuse union
drop if idcode>1000

* Fixed-effects BRGLM probit regression
brfeglm union age grade not_smsa, model(probit) identifier(idcode)

* Same as above, but with clustered standard errors
brfeglm union age grade not_smsa, model(probit) identifier(idcode) cluster(idcode) savef

*Summarise the fixed effects estimates
su __feidcode*

*Calculate average marginal effects
margins, dydx(*)

Update History

  • March 15, 2021
    • initial commit

Authors:

Alexander Ballantyne
University of Melbourne

Johannes Kunz
Monash University

Kevin E. Staub
University of Melbourne

Rainer Winkelmann
University of Zurich

References:

Bias reduced canonical link function models:

Firth, David. 1993. Bias Reduction of Maximum Likelihood Estimates. Biometrika. 80(1): 27-38.

Bias reduced generalised linear models:

Kosmidis, I., & Firth, D. 2009. Bias Reduction in Exponential Family Nonlinear Models. Biometrika, 96(4): 793-804.

Bias reduced fixed effect generalised linear models:

Kunz, J. S., K. E. Staub, & R. Winkelmann. 2021. Predicting Individual Effects in Fixed Effects Panel Probit Models. Journal of the Royal Statistical Society: Series A. 184(3): 1109-1145.

Published applications:

Buchmueller, T. C., Cheng, T. C., Pham, N. T., & K. E. Staub. (2021). The effect of income-based mandates on the demand for private hospital insurance and its dynamics. Journal of Health Economics, 75, 102403.

Kunz, J. S. & C. Propper. (2023). JUE Insight: Is Hospital Quality Predictive of Pandemic Deaths? Evidence from US Counties. Journal of Urban Economics, 133: 103472.

Kung C. S. J., Kunz, J. S. & M. Shields. (2023). COVID-19 Lockdowns and Changes in Loneliness among Young People in the U.K. Social Science & Medicine.320: 115692, 2023.

Krause, D. (2023). Armed Conflicts With Al-Qaeda and the Islamic State: The Role of Repression and State Capacity. Journal of Conflict Resolution, 0(0). In press. Replication files here.

Kunz, J. S., Propper, C., Staub, K. E. & Winkelmann, R. 2024. Assessing the quality of public services: For-profits, chains, and concentration in the hospital market Health Economics. In press.

About

Bias reduced fixed effect generalized linear models

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published