Logistic Regression Analysis Based on Jackknife Method: An Application on the Estimation of Coronary Artery Disease

Abstract

Coronary artery disease is a multifactorial disease characterized by various factors and their interactions, and is the most common cardiac disease. In this study, it is aimed to obtain the logistic regression model for the estimate of coronary artery disease and to compare the parameter estimates for the factors affecting the estimation of parameters to the parameter estimates for the logistic regression model based on the jackknife method. The parameter estimates are made by backward elimination method and a suitable regression model is determined. In the data analysis and parameter estimates, the R 3.3.3. and SPSS 23.0 package software programs are used. It is obtained that the classification percentages of the model are over 80%. When the results of the original logistic regression analysis and the logistic regression analysis based on the jackknife method are examined, it is seen that there is little difference between the coefficient estimates. It has been determined that the standard errors of the parameter estimates in the Jackknife method are lower than the standard errors of the parameter estimates obtained from the classical sample. As a result; it was found that the parameter estimates based on the Jackknife method are very effective and the logistic regression model based on this method was found to be quite successful for the estimation of coronary artery disease.

Keywords: Backward elimination, Jackknife, Coronary artery disease.