Modeling of Forecast Performance Indicators of Organizations

The aim of this research is to model forecast values based on performance indicators revealed and assessed using multivariate analysis. Forecast indicators of activity of organizations are modeled on the example of agro-industrial complexes of the Urals Federal District (UFD) in the livestock (pig-breeding) industry. Modern methodology requires conducting this work on the basis of the extensive use of information technology and economic and mathematical modeling. The basis for building models is the actual statistical information, which makes it possible to understand the real state and the possible development of the organization. In the construction of models, the identification of relationships, dependencies, and trends in economic indicators requires the determination and justification of the outcome indicator (dependent variable) and factor indicators (independent variables). In this study, the indicator “production of pigs for slaughter in live weight in agricultural organizations” was chosen as an outcome indicator and as factor indicators: the pig stock in agricultural organizations and the average daily gain on raising and fattening. We developed regression models that allowed the identification and assessment of correlations between the main production indicators of agricultural organizations on the basis of cluster analysis and panel data. It was revealed that cluster analysis made it possible to significantly increase the adequacy of regression models. The necessity of comparing the results of estimating standard regression models and taking into account variation within the observation object for different periods was justified.