Comparison Between Iterative Maximum Likelihood Estimators Method And Bayesian Method For Estimating Logistic Regression Model Parameters With Practical Application

Najm Abdallah, Zana; Ali Fakhri, Mustafa; Ali Chichan, Ali Mohammed

	Comparison Between Iterative Maximum Likelihood Estimators Method And Bayesian Method For Estimating Logistic Regression Model Parameters With Practical Application
journal of kirkuk University For Administrative and Economic Sciences
Volume 10, Issue 2, February 2020, Pages 259-270 PDF (1.51 M)
Authors
Zana Najm Abdallah; Mustafa Ali Fakhri; Ali Mohammed Ali Chichan
University of Kirkuk Journal For Administrative and Economic Science
Abstract
There are many conditions for using regression in general, sometimes the conditions of using regression are not fulfilled in this case we should find alternative methods of data analysis so that we can predict the phenomenon studied and the most important of these methods is the logistic regression method where logistic regression is one of the methods commonly used, especially in binary data, where we used in this research, two methods to estimate the parameters of the logistic regression model, a method iterative maximum Likelihood Estimators and Bayesian Method We used two criteria for comparison, which is the Mean Square Error, and the mean Absolute Percentage error. Real data was used in this research, which is represented by lung cancer with a sample size (30) taken from the City of Medicine Hospital / Cancer Hospital where the results showed that the method Bayesian the best by Comparison criteria in estimating logistic regression model parameters.
Keywords
logistic regression; iterative maximum likelihood Estimators; Bayesian; mean square error; mean absolute percentage error

Statistics Article View: 148 PDF Download: 64