Specification and estimation of random effects models with serial correlation of general form

This paper is concerned with maximum likelihood based inference in random effects models with serial correlation. Allowing for individual effects we introduce serial correlation of general form in the time effects as well as the idiosyncratic errors. A straightforward maximum likelihood estimator is derived and a coherent model selection strategy is suggested for determining the orders of serial correlation as well as the importance of time and individual effects. The methods are applied to the estimation of a production function for the Japanese chemical industry using a sample of 72 firms observed during 1968-1987. Empirically, our focus is on measuring the returns to scale and technical change for the industry.