Search

In the setting of the Stochastic Frontier Production Models, the productive efficiency of the 28 countries belonging to the European Union is analysed. To this end, panel data encompassing a broad period of time is selected, which facilitates the study into whether there is greater efficiency in the years of crisis as a consequence of the adjustment measures. A translog specification of a Cobb-Douglas model is considered, in which the output is measured through the GDP of the countries and two productive factors (capital and labour). The model also includes a trend component that addresses the possible presence of technological change, and dummy variables for each country in order to separate unobserved heterogeneity from productive inefficiency. In relation to the perturbation that models the inefficiency, a model of the type Battese and Coelli (1995) is considered with a trend component and a variable related to economic growth. Finally, the growth of productivity is decomposed into the sum of the changes in technology, in economies of scale, and in inefficiency.

The importance of a scientific journal is usually established by considering the number of citations received by the papers that the journal publishes. In this way, the number of citations received by a scientific journal can be considered as a measure of the total production of the journal. In this paper, in order to obtain measures of the efficiency in the production process, the approach provided by stochastic frontier analysis (SFA) is considered, and econometric models are proposed. These models estimate a frontier production, which is the maximum achievable number of citations to the journal based on its resources. The efficiency can then be measured by considering the difference between the actual production and the estimated frontier. This approach is applied to the measurement of the productive efficiency of the journals of the JCR social sciences edition database, which belong simultaneously to the areas of “economics” and “social sciences, mathematical methods”.

In this paper, the finite sample properties of the maximum likelihood and Bayesian estimators of the half-normal stochastic frontier production function are analyzed and compared through a Monte Carlo study. The results show that the Bayesian estimator should be used in preference to the maximum likelihood owing to the fact that the mean square error performance is substantially better in the Bayesian framework