home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math
- Path: sparky!uunet!usc!cs.utexas.edu!bcm!aio!gepr02!nas_ps
- From: nas_ps@geds01.jsc.nasa.gov (Pradip R. Shah)
- Subject: IMSL : RBEST ROUTINE
- Message-ID: <1992Dec14.195942.936@aio.jsc.nasa.gov>
- Sender: news@aio.jsc.nasa.gov (USENET News System)
- Reply-To: nas_ps@geds01.jsc.nasa.gov
- Organization: G.E. Government Services
- Date: Mon, 14 Dec 1992 19:59:42 GMT
- Lines: 21
-
- HI !!
-
- I Have the following problem with the IMSL (International Mathematical Statistical
- Library) software. Those of you who are familiar or are currently using it, I
- would sincerely appreciate your help.
-
- I am trying to generate a regression equation using the routine "RBEST" which selects
- the best multiple linear regression analysis. My problem is that I can not get
- the intercept. The design matrix, as I understand, should have the first column
- elements equal to 1 (for constant regressor). I am getting the singular matrix.
- The illustration uses the data from the Draper & Smith and I have been working with
- this data set before I can really analyze my data. If someone can help me with
- this problem, I would be grateful.
-
- Thanks.
-
- P.S. My reference is IMSL : Fortran routines for Statistical Analysis, Softcover
- edition 1.1, pgs. 176-182.
-
- Pradip Shah GE
-
-