To compare of two parameter-estimation methods. These two methods are Maximum Likelihood Estimation method (MLE) and Weighted Least Squares method (WLS). The model in this research uses identity link function In this study, the number of the independent variables is 1, 2 and 3. Sample size is 25, 50, 75, 100, 125, 150, 175, 200, 225, 250, 275, 300, 325, 350, 375 and 400. There are three levels of correlation among independent variables in the study. Those are no correlation, low correlation and high correlation. The data in this research are generated through Monte Carlo simulation using program s-plus 2000. The simulation was run 500 times for each situation. The AMSE (Average Mean Square Error) is used as the evaluation criterior for both methods. The conclusion is that Maximum Likelihood Estimation method and Weighted Least Squares method provided almost the same estimates and no one is better than the other significantly. At the different levels of correlation among independent variables, it can not be concluded clearly which level gives the better estimates.