Production analysis: A non-parametric time series application to US agriculture

The paper investigates production analysis using non-parametric methods. When all data are not consistent with profit maximisation, we propose a method relying on the concept of 'effective inputs and outputs' to evaluate technology and economic behaviour. The approach is illustrated in a time series application to US agriculture. The results provide empirical measures of productivity growth. They also generate useful information on the 'bias in technical change' in US agriculture.