- All Journals
- Al-Qadisiyah University
- Basrah University
- Kirkuk University
- Mosul University
- The Iraqi Borad for Medical Specialization
- University of Anbar
- University of Technology
- Al Mustafa University
- Al-Zahraa University for Women
- Alnoor University
- Kunooz University
- Misan University
- Ninevah University
- Tikrit University
- University of Al-Hamdaniya
- University of Kerbala
When and why fail stepwise way in choosing the best subset Variables for regression | ||
| Journal of Administration and Economics | ||
| Article 1, Volume 0, Issue 95, 2018, Pages 342-351 | ||
| Author | ||
| A. M. D. Sabah Abdul Hussein Faraj | ||
| Abstract | ||
| No doubt that step wise procedure (SP) is one of the best procedures of serial choice of regression variables, but it suffers – as I see – from abasic defect that, in most applications it can't getrid of the effect of first selected variable on selecting the other ones in case of presence of linear correlations between them. The first selected variable stands as a screen before joining the other important variables "totally or partially" the finalsubset. Consequently, (SP) fails selecting the best subset regression variables (BSRV). So there will be no use of retesting it during the procedure unless one breaks the screen in such away like raising the significance level of testing the selected variables such that it won't be less than 0.10 and according to their importance. This is what the researcher has proposed to make the (SP) more efficient than foreword to select (BSRV). But one must notice that this proposal may lead to select additional variables that have weakened contribution in explanation of social phenomenon. | ||
| Keywords | ||
| Statistical axis | ||
|
Statistics Article View: 20 PDF Download: 1 |
||