Regression Analysis of an Experiment with Treatment as a Qualitative Predictor

Abstract

This research work explores the indept use of multiple regression analysis in modelling the optimum weight gain of Broiler chicken as a function of initial weight and feed combination (treatment). The statistical relationship between the weight gain, initial weight and different feed combination was established using a regression model. The measurement on the initial weights was kept silent, allowing the use of analysis of variance technique to determine the error inherent in the study. Application of the model was done on the result of the experiment carried out on agritted breed of Broilers. The best feed combination (among the different feed combination considered) was determined by looking out for the combination that yields the optimum weight gain when the initial weight is known.