Correct and Incorrect Error Specifications in Statistical Cost Models

This paper reports on a Monte Carlo study of the accuracy of each estimation method under correct and incorrect error specifications. The multi-predictor model that serves as the basis of the study is representative of models with which cost analysts frequently deal. Results suggest that ordinary least squares, applied to the logs of the variables, may be the preferred method under either specification. This finding, however, is applicable strictly to the issue of parameter estimation. Accuracy of cost prediction requires further analysis.