Some experiences applying fuzzy logic to economics

Barbara Díaz

Antonio Morillas

pp. 347-379

in: Rudolf Seising, Veronica Sanz González (eds), Soft computing in humanities and social sciences, Berlin, Springer, 2012

Abstract

Economy becomes a field of special interest for the application of fuzzy logic. Here we present some works carried out in this direction, highlighting their advantages and also some of the difficulties encountered. Fuzzy inference systems are very useful for Economic Modelling. The use of a rule system defines the underlying economic theory, and allows extracting inferences and predictions. We applied them to modelling and prediction of waged-earning employment in Spain, with Jang's algorithm (ANFIS) for the period 1977-1998.As additional experiences in this direction, we have applied the IFN algorithm (Info-Fuzzy- Network) developed by Maimon and Last to the study of the profit value of the Andalusian agrarian industry.The search for key sectors in an economy has been and still is one of the more recurrent themes in Input-Output analysis, a relevant research area in the economic analysis. We proposed a multidimensional approach to classify the productive sectors of the Spanish Input- Output table. We subsequently analyzed the problems that can arise in key sector analysis and industrial clustering, due to the usual presence of outliers when using multidimensional data.