A Multivariate Analysis on the response of crop to fertilizer and soil type

Abstract

This research work is primarily aimed at determining the significance effect of factors and to test the determinants of farmers participating behavior in agricultural. A model of Multinomial Logistic Regression (MNL) is employed and other method that have been chosen which are Principal Component Analysis (PCA), and Factor Analysis (FA). Using multinomial regression the dependent variable in question is a nominal where more there are more than two categories [1], analyze of series of data which observations are explained by variance [2], and to find simple patterns in the pattern of interactions between variables. Analysis of variance using PCA and FA and the pattern to eliminate the effect of those factors accordingly after the data analysis are concluded from the analysis of variance that replicates soil types is individually individually main effect statistically significance at 5% significance level since P-value 0.002 which below 0.05. The soil type achieved the highest variance of 50.1% compare to fertilizer type with 49.9% for being a factor and response for a crop also, interaction effect of fertilizer and replicates is  statistically not significance since the P-value below 0.05 accordingly.