The technique of economic efficiency estimation of in-tegrated vision of the sugar industry AIС

Abstract

Purpose: the evaluation and verification of the mathematical apparatus is an integral part of any research. The use of a real statistical base in the constructed models allows us to evaluate their adequacy, accuracy of approximation, and predictive ability. This is confirmed by the consistency of the theoretical results with the experimental ones. The article deals with the issues of verification of complex forecasting of performance indicators of integrated production systems of the sugar subcomplex (IPS SS), taking into account their industry features. The subject of the study is the confirmation of the proposed scientific hypotheses and propositions on the basis of a thorough analysis of the state of research in this subject area. Discussion: the authors suggest that the adapted and new metrics of verification of forecasting systems significantly improve the quality of the forecast, which is defined as a measure of the sustainable development of the object according to the trajectories that were determined by the forecast plan. The author proposes to perform complex verification for all studied objects using the following tests: crossvalidation method, predicativity coefficient, Mann – Whitney test, detrend fluctuation analysis, V-statistics, scaled window variance method, white noise tests, homogeneity tests, tests of historical profitability of predictions, coherence coefficient, graphical analysis of models, etc. For the first time, verification tests are implemented: forward efficiency coefficient and quantitative indicator of system quality. It should be remembered that certain provisions require additions, clarifications or the involvement of other information and developments, which increases their methodological usefulness. Results: the article describes a number of limitations and difficulties that researchers engaged in this scientific field will have to overcome. The author outlines the prospects and ways of further development of verification of complex forecasting not only for the sugar industry, but also for other sectors of the national economy of Russia.