Improper problem of forecasting basedon heterogeneous information

Abstract

The paper considers the problem of forecasting based on heterogeneous information, i. e. the statistical data and expert judgments represented in the form of inequalities that form an inconsistent system. To solve the studied problem, it is proposed to reduce it to an improper linear programming problem and use one of the existing methods of matrix correction. The correction of the linear programming problem is considered as a special case of the multi-objective optimization problem.