Identification of parameters of a combined piece-linear regression model

Abstract

Subject of research: the problem of estimating the parameters of a combined piecewise linear regression model.
 Purpose of research: to apply the apparatus of linear Boolean programming to identify its parameters. 
 Methods and objects of research: the object of research is a set of indicators of inventive activity in the Russian Federation, the methods are regression analysis and mathematical programming.
 Main results of research: an approach to estimating the parameters of a combined piecewise linear regression model by using the method of least modules is described, which makes it possible to reduce this problem to a linear Boolean programming problem. A combined piecewise linear model of inventive activity in the Russian Federation has been constructed, which can be used in various analytical and predictive calculations. The output variable of the model is the number of patent applications, and the input variable is the volume of gross domestic product, the number of graduate students and teaching staff.