Serial correlation estimation through the imprecise Goal Programming model

The Goal Programming (GP) model was used as a time-series analysis tool that incorporates a Serial Correlation where the dependent variable is considered as precise. This formulation does not take into consideration the decision-maker's preferences. However, the dependent variable can be imprecise and its value can be expressed through an interval. The aim of this paper is to develop a new formulation of the GP model for regression with Serial Correlation where the dependent variable is imprecise. The proposed model will also integrate explicitly the decision-maker's preferences. A numerical example was used to illustrate our model.