Regression model of valerian root yields in the Forest-Steppe zone of Ukraine depending on fertilization rates and water use of the crop

Abstract

The study is devoted to the creation and description of the mathematical model of valerian root productivity formation in the Forest-Steppe zone of Ukraine. The study is based on the perennial (2013-2015) research results of Ukrainian scientists in the field of medicinal plant cultivation in the drip-irrigated conditions. The generalization and statistical analysis of the data was performed, and the multiple linear regression model of valerian root yields was proposed. The results of modeling testify that the effect of mineral fertilizers is greater than the latter of water supply for the crop in the above-mentioned conditions of growing. At the same time, it was proved that the value of irrigation is sufficient, and the crop requires adequate irrigation to form high yields of root mass. The figures of the coefficient of determination for all the models, engaged in the study, fluctuate within 0.96-0.99, testifying about their high descriptive value. The mean absolute percentage errors of the models are within 10-18%, therefore, their performance is good enough for operational planning and practical use in the crop yield prediction depending on the studied agro-technological factors.