THE USE OF FUZZY LINEAR REGRESSION IN THE MODEL OF TECHNOLOGICAL KNOWLEDGE GROWTH

Abstract

Fuzzy linear regression is used to estimate the parameters of the Botazzi-Peri model of technological knowledge growth, while classical linear regression is used to evaluate the parameters. In this paper, the parameters are defi ned as fuzzy values. As a result, the model volume of technological knowledge is characterized by a set of possible values rather than by a single number. The position in a series of possible values for a particular indicator characterizing a country, to some extent, refl ects the effectiveness of R & D in this country. If the fi gure is close to the most probable values of the model (when the value of the membership function is close to one), the effectiveness of R & D is quite typical. Closeness to peripheral values (when the value of the membership function is close to zero) occurs in two cases: if the effi ciency of R & D is atypically high or atypically low. The use of fuzzy regression allows to trace the dynamics of changes in effectiveness. The calculations were performed using the same data on body of technological knowledge in developed countries as in Botazzi and Peri work.

Keywords

• fuzzy linear regression
• growth model
• body of technological knowledge