Comparing means of two non-homogeneous normal populations
The researchers often avoid the maximum likehood (M.L) procedure that is known to have a number of desirable properties, because the solutions are complicated. However, it can be attractive to many if some efficient algorithm is available. This paper intends to give an alternative approach for testi...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universiti Utara Malaysia
1986
|
| Subjects: | |
| Online Access: | http://repo.uum.edu.my/117/1/Mohd_Nawi_Abd_Rahman.pdf http://repo.uum.edu.my/117/ http://ijms.uum.edu.my |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The researchers often avoid the maximum likehood (M.L) procedure that is known to have a number of desirable properties, because the solutions are complicated. However, it can be attractive to many if some efficient algorithm is available. This paper intends to give an alternative approach for testing the means of two normal populations having unequal variances but whose coefficient of variations are homogeneous. The M.L. method is appropriate. The statistic is a funtion of the three M.L. estimates. A simple algorithm which can be implemented on a microcomputer is proposed for calculating the required values. The statistic is shown to be a one-degree chi-square
variable. An application on the proposed technique using an
experimental data is given. |
|---|