# Prediction of the number of domestic animals and birds in the conditions of the economic crisis

Andre Nepochatenko^{1}, Victor Nepochatenko^{2}, Uliana Revitska^{2}, Oksana Strigina^{2}, Olena Melnichenko^{2} and Vitaly Bomko^{3}

^{1}*Bila Tserkva National Agrarian University, Faculty of Economic, Department of Entrepreneurship, Trade and Exchange Activity, Bila Tserkva, 09117 Ukraine*

^{2}*Bila Tserkva National Agrarian University, Faculty of Economic, Department of Higher Mathematics and Physics, Bila Tserkva, 09117 Ukraine*

^{3}*Bila Tserkva National Agrarian University, Faculty of Biological Technology, Department of Technology of Feed, Feed Additives and Animal Feeding, Bila Tserkva, 09117 Ukraine*

### Abstract

Nepochatenko, A., Nepochatenko, V., Revitska, U., Strigina, O., Melnichenko, O. & Bomko, V. (2020). Prediction of the number of domestic animals and birds in the conditions of the economic crisis. *Bulg. J. Agric. Sci*., 26 (4), 731–736

The article analyzes the nonlinear regressions, on the basis of which the forecast of economic characteristic in the crisis period is made. Regression functions are obtained by solving the corresponding differential equations. It is shown on the example of the change in the number of animals in Ukraine and Russia for 1990–2017, what these changes may correspond to an exponential, logistic regression or their modifications. It is shown that in the crisis period, the dynamics of the decrease in the number of livestock corresponds to the modified exponential regression. It is suggested to find the two parameters of these regressions using the least squares method, the third parameter to be determined by the numerical method with the minimum average absolute percentage error (MAPE). At the exit of the crisis state, when the process of increasing the number of livestock begins, the dynamics corresponds to a modified logistic regression. Two logistic regression parameters were determined using the least squares method, the third and fourth parameters calculate by a numerical method from the condition of a minimum MAPE as a function of two variables. The theoretical findings obtained are in good agreement with the statistics corresponding to the dynamics of cows, pigs, sheep, goats and poultry in Ukraine for the period 1990–2017.

*Keywords*: prediction; exponential and logistic regressions; time series forecasting