Forecasting the effectiveness of land use in agricultural enterprises

Abstract

The problems of forecasting the efficiency of land use in the production of sunflower, corn for grain and wheat in enterprises of the Kharkiv region are studied. The purpose of the study is to build predictive models of the dependence of productivity on the amount of costs for individual items in the production of sunflower, wheat and corn for grain in agricultural enterprises of the Kharkiv region according to 2020 data. At the first stage of the research, those expenditure items that had a reliable level of connection with the level of crop productivity were checked and selected. The criterion in this case was the t-Student value. The constructed regression model of the dependence of the yield value of the selected crops on the level of expenditure on individual articles was reliable, but had an average closeness of connection. Checking the level of adequacy of the built models from the point of view of their use for forecasting made it possible to establish that the average and median values of the actual and forecast yields of all crops have close values. The values of the maximum and minimum values differed significantly. It is proposed to determine the model coefficient of land use efficiency. The economic meaning of this coefficient is that if its value is greater than 1, it will mean that the vast majority of enterprises in this population had a higher crop yield than it should have based on the average level of use of spent resources. On the basis of actual data on each crop and its specific weight in the structure of land use, the final value of the model coefficient of land use efficiency was obtained on the example of agricultural enterprises of the Kharkiv region. The value of the proposed coefficient was equal to 0.868. It is concluded that most of the enterprises had a level of efficiency of land use according to the criterion of crop yield lower than could be expected based on the spent resources.