The paper proposes a method for solving the problem of simultaneous maximization of the dynamic criterion for the consistency of the behavior of the calculated and actual values of the dependent variable of the linear regression model and minimization of the loss function corresponding to the method of least absolute deviations. This method allows one to reduce such a problem to a partially Boolean linear programming problem.

linear regression model, loss function, method of least absolute deviations, dynamic criterion of behavior consistency, partial Boolean linear programming problem