Prediction in the one‐way error component model with serial correlation

This paper derives the best linear unbiased predictor for a one-way error component model with serial correlation. A transformation derived by Baltagi and Li (1991) is used to show how the forecast can be easily computed from the GLS estimates and residuals. This result is useful for panel data applications which utilize the error component specification and exhibit serial correlation in the remainder disturbance term. Analytical expressions for this predictor are given when the remainder disturbances follow (1) an AR(1) process, (2) an AR(2) process, (3) a special AR(4) process for quarterly data, and (4) an MA(1) process.

[1]  L. Lillard,et al.  Components of variation in panel earnings data: American scientists 1960-70 , 1979 .

[2]  T. Kloek,et al.  On a simple transformation for second-order autocorrelated disturbances in regression analysis , 1973 .

[3]  A. Goldberger.  Best Linear Unbiased Prediction in the Generalized Linear Regression Model , 1962 .

[4]  T. J. Wansbeek,et al.  The separation of individual variation and systematic change in the analysis of panel survey data , 1978 .

[5]  Thomas MaCurdy,et al.  The use of time series processes to model the error structure of earnings in a longitudinal data analysis , 1982 .

[6]  Badi H. Baltagi,et al.  A transformation that will circumvent the problem of autocorrelation in an error-component model , 1991 .

[7]  Marc Nerlove,et al.  Pooling Cross-section and Time-series Data in the Estimation of a Dynamic Model , 1966 .

[8]  Allan J. Taub.  Prediction in the context of the variance-components model , 1979 .

[9]  P. Balestra.  A note on the exact transformation associated with the first-order moving average process , 1980 .

[10]  Robert J. Willis,et al.  Dynamic Aspects of Earning Mobility , 1978 .

[11]  Tom Wansbeek,et al.  A simple way to obtain the spectral decomposition of variance components models for balanced data , 1982 .

[12]  T. Wansbeek,et al.  A Note on Spectral Decomposition and Maximum Likelihood Estimation in ANOVA Models with Balanced Data , 1983 .

[13]  J Lew Silver,et al.  Generalized Estimation of Error Components Models with a Serially Correlated Temporal Effect , 1982 .