Prediction of the lean meat content of carcass with multivariate regression method in Guanzhong Black Pig

Thirty Guanzhong Black pigs with a live weight about90 kg each were selected in this study.The live weight (X1) ,final body length (X2 ) ,dressed weight (X3) ,carcass length (X4 ) ,eye area (X5) ,ham joint weight (X6) ,back- fat thickness (X7) were used as independent variables,and the lean meat weight of carcass (Y) was used as a dependent variable,using the method of stepwise regression analysis.After statistical examination and omitting the small insignificant independent variables,an optimum three- way regression equation for estimating the lean meat weight was established with R= 0 .9575:Y=- 30 .51 94+0 .5537X3+0 .2 742 X4 +0 .331 6X5,in which X3refers to the dressed weight (kg) ,X4 refers to the carcass length (cm) ,X5refers to the eye area (cm2 ) .The average value of prediction of the lean meat weight using the equation and the observed value were proved to have no significant difference through the t- test(P0 .0 5) in43Guanzhong Black Pigs.Itshows this method is a simple and easy way to predictthe lean meatcontentin pork carcass accurately,and can be used in both research and production.