Comparison of methods to estimate a multiple linear regression model when the error is distributed Platykurtic
Abstract
In this paper, we used three methods (Maximum Likelihood Method, Modified Maximum Likelihood Method, Bootstrap Method) to estimate the parameters of the multiple linear regression model when the random error is distributed (Platykurtic distribution) and the results show that the modified maximum likelihood method is the best after comparing between the methods by Using mean square error for the model and the parameters and four sample sizes (20,60,100,200) and by simulation.