The Comparison of Point Estimation for Parameter for Geometric Distribution Data in Small Sample Size

บทคัดย่อ

Abstract

The objective of this research is to compare two point estimation methods: Maximum Likelihood Method (MLE) and Bayesian Method (Baye). When data is Geometric distribution, the parameters (p) are 0.1, 0.3, 0.5, 0.7 and 0.9, whereas the sample sizes (n) are 3, 5, 8, 10, 12, 15, 18, 23, 25, 28 and 30. In each situation, the data has been simulated and repeated for 1,000 times. The Mean Absolute Error is used as a criterion for comparison.

According to the results, when the sample sizes are 3, 5, 8 and 10, on overall MLE yields the least mean absolute error when parameter equal to 0.1. Whereas parameter larger than or equal to 0.3, on overall Baye yields the least mean absolute error. when the sample size equal to 12, 15, 18, 20, 23, 25, 28 and 30, on overall MLE yields the least mean absolute error when parameter equal to 0.1 and 0.3. Whereas parameter larger than or equal to 0.5,on overall Baye yields the least mean absolute error.

Keywords: Geometric Distribution, Point Estimation Introduction In