Proposed Generalized Method and Algorithms for the Estimation of Parameters and Best Model Fits of Log Linear Model

Abstract

The paper is on proposed generalized method and algorithms developed for estimation of parameters and best model fits of log linear model for q-dimensional contingency table. For purpose of this work, the method was used to provide estimates of parameters of log –linear model for five- dimensional contingency table. In estimating these parameters and best model fit, computer programs in R were developed for the implementation of the algorithms. The iterative proportional fitting procedure was used to estimate the parameters and goodness of fits of models of the log linear model.  A real life data was used for illustration and the result obtained showed the best model fit for five dimensional contingency table is [BSGM, BGAM]. This showed that the best model fit has sufficient evidence to fit the data without loss of information. This model has highest p-value and the least likelihood ratio estimate. This model also revealed that state of origin is independent of age given Bscgrade and mode of admission.

Keywords:  contingency table  , categorical data, hierarchical log –linear models, Parameters, proposed generalized  method, algorithms,  Iterative proportional fitting  procedure.