This paper is concerned with the estimation of production functions and measurement of the rate of technical change. Multiple time trends are introduced as an alternative to the single time trend representation of technical change. The underlying technology is represented by Cobb-Douglas and translog functional forms. Random effects models with homoscedastic variances is assumed. The models are estimated using the generalized least squares method. The data used are a large rotating panel data set from Swedish crop producer farms over the period 1976-1988. The empirical results show that a single or multiple time trends representation yield different time behaviour of technical change. The latter is found to perform much better.