Using mixtures of flexible functional forms to estimate factor demand elasticities

Researchers who wish to estimate factor demands using flexible functional forms may now choose from several candidates supplied by the theoretical literature. Unfortunately, the criteria for a priori model selection are not clear. This paper adopts the use of Bayesian methods and argues that there is in fact no need to choose; the optimal strategy is to use a mixture of functional forms to estimate the parameters of interest. Problems of overfitting are avoided by the imposition of the appropriate regulatory conditions. Practical implementation is greatly simplified by the use of Markov chain Monte Carlo techniques. In an example, three well-known functional forms for cost functions are applied to estimate factor demand elasticities in the Canadian manufacturing sector.