USING MULTILOMIAL LOGISTIC REGRESSION TO PREDICT NITRATE NITROGEN CONTENT IN SOIL
2022
Kizimova, T.A. (Siberian Federal Scientific Centre of Agro-BioTechnologies of the Russian Academy of Sciences)
This paper presents a multinomial logistic regression model that allows predicting and thereby quickly determining the content of nitrate nitrogen in the 0-40 cm soil layer before sowing. To build and train the model the data of a long-term multifactorial stationary field experiment of the Siberian Research Institute of Husbandry and Chemicalization of Agriculture (SRIHCA) of the Siberian Federal Scientific Centre of AgroBioTechnologies (SFSCA) of the Russian Academy of Sciences (RAS) of the time period of 2009 -2018 were used. During the analysis of the data sample (observations), the main predictors of the model that affect the content of nitrate nitrogen in the soil (target indicator) were identified. The predictors are represented by qualitative and quantitative parameters of the working area: predecessor, tillage, weather conditions, productive moisture content in the soil before sowing, nitrate nitrogen content by appropriate gradations. The quality of the developed multinomial logistic regression model was assessed using the coefficient of determination, which was 78% according to the Nagelkerke measure, and 72% according to the Cox - Snell measure. The predictive capabilities of the trained model were evaluated. The overall proportion of correct predictions for the multinomial logistic regression is 80.6%.
显示更多 [+] 显示较少 [-]