Non-Extensive Entropy Econometric Model (NEE): the Case of Labour Demand in the Podkarpackie Province

Abstract

The non-extensive entropy (NEE) principle has been successfully applied in the case of high frequency financial market analysis. I try to extend the approach to empirical social sciences and propose a competitive estimation approach with respect to classical econometrical methods. This article constitutes a limited extension of Jaynes‐Shannon‐Gibbs’ (JSG) ergodic system formalism already applied to classical econometrics. The Podkarpackie private labour demand model is then developed and its outputs presented. A constrained weighted dual criterion function maximising entropy probabilities for parameter and disturbance components is derived and its inferential information indexes are proposed and computed. We note that the increase of relative weights on disturbance component leads to higher values of q, the entropic index of generalized Tsallis entropy. Smaller disturbance weights produce q values closer to unity. Outputs then converge to those displayed by the competitive JSG and least squares (LS) approaches. However, finding out an inferential rule delimiting the critical q values for Gaussian distribution interval remains of high interest. In terms of economics, the results of the proposed model show a realistic adjusting speed mechanism of actual lever of employment to its long run targeted equilibrium level owing to expected market profits.