Center for Policy Research
Working Paper
Estimation of Serially Correlated Error Components Models Using Whittle’s Approximate Maximum Likelihood Method
Badi H. Baltagi, Georges Bresson, and Jean-Michel Etienne
C.P.R. Working Paper No. 271
September 2025
Abstract
This chapter studies the estimation of error components models with serial correlation of the 𝐴𝑅𝑀𝐴(𝑝, 𝑞) type using Whittle’s (Whittle, 1953) approximate maximum likelihood method. This is done for the one-way and two-way error components panel data model. Monte Carlo simulations are performed that investigate the small sample performance of this method.