**Date:** Fri, 4 Oct 2002 07:29:01 -0400
**Reply-To:** Raynald Levesque <rlevesque@videotron.ca>
**Sender:** "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
**From:** Raynald Levesque <rlevesque@videotron.ca>
**In-Reply-To:** <HCEDIJNOJNFAPGOJOMFGCEONCAAA.tech@spss-sa.com>
**Content-type:** text/plain; charset=iso-8859-1
Hi

This adresses your question 2) below.

The following is a simple example:

GET FILE="C:\Program Files\SPSS\employee data.sav".
REGRESSION
/MISSING LISTWISE
/STATISTICS COEFF OUTS R ANOVA
/CRITERIA=PIN(.05) POUT(.10)
/NOORIGIN
/DEPENDENT salary
/METHOD=ENTER educ jobtime prevexp
/OUTFILE=COVB('C:\Temp\OLS param.sav') .

The above syntax saves the following info to the spss file "OLS
param.sav":
- covariance matrix of the 3 IV's
- coefficients of the model
- SE of the coefficients
- SIG and Degrees of Freedom

That file can then be loaded back in SPSS or exported to text file by
SPSS.

THis is the content of the file "OLS param.sav":
depvar_ rowtype_ varname_ const_ educ jobtime prevexp
SALARY COV CONST_ 30575136 -578502 -268251 -6972.12
SALARY COV EDUC -578502 44364.28 -612.27 308.98
SALARY COV JOBTIME -268251 -612.27 3415.30 -5.24
SALARY COV PREVEXP -6972.12 308.98 -5.24 33.68
SALARY EST -27886.3 4004.58 87.95 11.94
SALARY SE 5529.48 210.63 58.44 5.80
SALARY SIG .00 .00 .13 .04
SALARY DFE 470.00 470.00 470.00 470.00

For a more complicated example, see "Add casewise regression coefficients
to data file.SPS" in

http://pages.infinit.net/rlevesqu/SampleSyntax.htm#RegressionRepeatedMeasure
s

In case a procedure does not export a given element of the results into a
matrix or data file, it is always possible to write a script which picks
up that information and save it to a text file.

HTH

Raynald Levesque rlevesque@videotron.ca
Visit my SPSS Pages http://pages.infinit.net/rlevesqu/index.htm

-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU]On Behalf Of
Tasneem Jappie
Sent: October 4, 2002 5:12 AM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject:

Hi,

I'm looking for relationships between measured values that we obtain during
our manufacturing of optical fiber. I believe that there is a relationship
between the Waterpeak, 1550 and 1310 values. There might be other
relationships that I don't know of, and would like to know if they exist.

I would like to model the results of one of the two measuring machines:
1)Can SPSS link with SQL database and on a time basis do linear regression
analysis?

2) I would like to use the coefficients generated by the regression, in a
Visual Basic Program. Would it be possible to extract the coefficients out
of SPSS? Does SPSS generate a Database file that I can open to extract the
coefficients?

Thank you.

Tasneem Jappie
Technical & Training Support
SPSS-SA
Silvermine House
Steenberg Office Park
Tokai 7945
South Africa
Tel +27 21 702 4666
Fax +27 21 702 4333
E-mail: tech@spss-sa.com
Web: www.spss-sa.com