Comparison of Linear and Non-Linear Models for Definition of Body Features of Romanov Lambs

Abstract

In this study, the purpose of defining the development of body characteristics of 50 Romanov lambs 180. During the growth period up to the age of day, records of body characteristics such as body weight, body length, height of cidago were taken every 15 days and the parameters of the growth curves were calculated from the Linear models with the obtained data and the Linear, Quadratic and Cubic model, Non-Linear models with Gompertz, and Logistic models. The coefficient of determination (R2), mean square error (MSE) and mean absolute deviation (MAD) values were used in determining the model that best fit the growth curve. As a result of the study, the highest R2 value and the lowest HKO values were 0.992-0.591 in live weight, 0.993-0.441 in cidago height, and 0.986-1.164 in body length, respectively. The highest R2 value in all body characteristics was obtained from the cubic model. SPSS statistics program was used to determine the parameters of the growth curve model. According to the obtained data, it was determined that the most compatible model to explain the development of all body characteristics of the Romanov lambs is the Cubic model.