Prediction of Egg Weight from External Egg Traits of Guinea Fowl Using Multiple Linear Regression and Regression Tree Methods

The study was done to predict egg weight from the external traits of Guinea fowl using statistical methods multiple linear regression (MLR) and tree analysis (RTA). A total 110 eggs a flock 23-week-old were evaluated. Egg (EW) traits: eggshell (ESW), polar diameter (EPD), equatorial (EED), shape index (ESI), surface area (ESA) measured. Descriptive statistics, Pearson correlation coefficients, equations MLR obtained; additionally, RTA CHAID algorithm with SPSS software (IBM ver. 22). EW presented positive correlations (p<0.0001) ESA (r = 0.72), EPD 0.65), EED 0.49). can be predicted through as predictor variable (R2 72%). Predictive accuracy improves when adding model 75%). built diagram ESA, EED, significant independent variables; these, most important (F 50,295, df1 4, df2 105; Adj. p<0.000) variation explained for 74%. Likewise, showed that highest (41.818 g) is obtained > 59.03 cm2 5.10 cm. proposed used reliably fowl.