USING FRACTIONAL REGRESSION MODELS TO ANALYSE EFFICIENCY OF EUROPEAN MANUFACTURING FIRMS

Abstract

Fractional regression models are effective when the variable of interest appears in the form of proportion or fraction or rate. In other words, use of fractional regression models becomes appropriate when the dependent variable is defined only on the standard unit interval. This paper surveys fractional regression models, and studies efficiency scores of manufacturing firms of Portugal, Greece, France, and Germany by using data from the years 2007 and 2013. The study consists of two stages. The first stage uses data envelopment analysis (DEA) to obtain efficiency scores. Then the second stage employs fractional regression models to analyze those scores. The main aims of the second study are to find how financial crisis, use of technology (High-Tech or Low-Tech), firm size (SME or large), and country of origin affect efficiency of firms. Results provide evidence that financial crisis does not affect efficiency of firms, use of technology does not play any role in explaining efficiency of firms, large firms more efficient than SMEs, and firms in Greece are comparatively less efficient.