Budi, Pratikno and Yuliatri , Wirawidya Haryono (2013) PENGUJIAN INTERCEP UNTUK TESTS TERKAIT NON-SAMPLE PRIOR INFORMATION PADA HIPOTESIS SATU ARAH PADA REGRESI LINIER SEDERHANA KETIKA VARIANSI DIKETAHUI. Prosiding Seminar Nasional Matematika dan Pendidikan Matematika, "Penguatan Peran Matematika dan Pendidikan Matematika untuk Indonesia yang lebih Baik". ISSN 978 – 979 – 16353 – 9 – 4
|
Text
S - 5.pdf Download (196kB) | Preview |
Abstract
This research discussed testing the intercept with non-sample prior information (NSPI) on the slope on a simple linear regression for known variance. The hypothesis used is one-side hypothesis (maximum), and the tests are unrestricted test (UT), restricted test (RT), and preliminary-test test (PTT). The method for chosing the best choice of the tests is a maximum power and minimum size. A simulation study and graphical analysis are given to make comparison of the tests. The result shown that the NSPI reduce the value of power, it means that the NSPI is not significant influence. However, PTT still lies between UT and RT for several intervals of 1, so it will be a better choice than UT and RT. Keywords: Linear model, power, and size.
| Item Type: | Article |
|---|---|
| Subjects: | Prosiding > Seminar Nasional Matematika dan Pendidikan Matematika 2013 |
| Divisions: | Fakultas Matematika dan Ilmu Pengetahuan Alam (FMIPA) > Pendidikan Matematika > Pendidikan Matematika |
| Depositing User: | Jurusan Pendidikan Matematika |
| Date Deposited: | 10 Dec 2013 03:28 |
| Last Modified: | 10 Dec 2013 03:28 |
| URI: | http://eprints.uny.ac.id/id/eprint/10824 |
Actions (login required)
 |
View Item |