Search

The stochastic frontier analysis (Aigner et al., 1977, Meeusen and van de Broeck, 1977) is widely used to estimate individual efficiency scores. The basic idea lies in the introduction of an additive error term consisting of a noise and an inefficiency term. Most often the assumption of a half-normal distributed inefficiency term is applied, but other distributions are also discussed in relevant literature. The natural estimation method seems to be Maximum Likelihood (ML) estimation because of the parametric assumptions. But simulation results obtained for the half normal model indicate that a method of moments approach (MOM) (Olson et al., 1980) is superior for small and medium sized samples in combination with inefficiency not strongly dominating noise (Coelli, 1995). In this paper we provide detailed simulation results comparing the two estimation approaches for both the half-normal and the exponential approach to inefficiency. Based on the simulation results we obtain decision rules for the choice of the superior estimation approach. Both estimation methods, ML and MOM, are applied to a sample of German commercial banks based on the Bankscope database for estimation of cost efficiency scores.

The paper explores the investment behaviour of German firms in the context of the Qapproach, which plays a dominant role in empirical investment research. The analysis is based on the Deutsche Bundesbank's corporate balance sheet statistics. The panel data set contains some 2,300 German firms' balance sheet data covering the years 1988-1998. While the Q-theory is mainly applied on the basis of stock market data, which facilitates the exploitation of market expectations and the calculation of average Q, the direct forecasting approach (Chirinko 1993) suggested by Abel and Blanchard (1986) and extended to panel data by Gilchrist and Himmelberg (1995, 1998) enables the Q-theory to be applied to non-quoted firms which are by far the majority in Germany. One of the key variables when using balance sheet data, which has attracted much detailed research, is firms' net capital stock at replacement costs. The challenge is to transform historical cost data, depreciated at non-economic, tax-oriented depreciation rates, into unreported and probably unknown economically meaningful data at actual replacement values. We suggest a complex procedure for calculating reliable replacement values of a firm's capital stock. To calculate Q we follow two different operationalisation strategies. First we estimate average Q based on balance sheet data by forecasting the present value of future profits using a VAR model. Second, we estimate marginal Q following the approach suggested by Gilchrist and Himmelberg. We compare the results from two different estimation techniques for dynamic investment models, GMM and direct bias correction. The results show that marginal as well as average Q influence investment significantly. When classifying the firms by size, we find that smaller firms react more strongly to Q and, to a lesser extent, to lagged investment.

Dissertation, Universität Duisburg-Essen, 2017 (Kumulative Dissertation)