| T O P I C R E V I E W |
| MicheleFerrari |
Posted - 10/25/2005 : 08:48:31 AM
I have some problems with my labtalk script doing multiple regression with the stat object.
My worksheet "Data1" includes 3 columns (p,b,c) with 18 rows. I would like to perform a multiple regression with "p" as dependent and "b" and "c" as independent variables with the following labtalk script:
stat.reset(); //Reset the DLL
stat.mr.YData$=Data1_p; //dependent variable
stat.mr.xData1$=Data1_b; //indep. variable b
stat.mr.xData2$=Data1_c; //indep. variable c
stat.mr();
//for the output some parameters
stat.mr.NumX=;
stat.MR.B1=; //parameter b1
stat.MR.B2=; //parameter b2
stat.SSR=; //y-intercept
Although multiple regression via menu (statistics-->multiple regression) gives the results (y-intercept = 917.8; b=-1.65; c=1.01 etc.), my labtalk script indicates all parameters = 0 (stat.MR.a, stat.MR.B1, stat.MR.B2), except the stat.mr.NumX = 2 (corresponding to the total number of X columns).
Could anyone help me with the script and tell me whats wrong?
Origin Version (Select Help-->About Origin): 7.5
Operating System: WIN2000 |
| 2 L A T E S T R E P L I E S (Newest First) |
| MicheleFerrari |
Posted - 10/26/2005 : 02:36:47 AM
Indeed it works! Thought I had already tried this out, but seems not to be the case.
Thank you very much! |
| Hideo Fujii |
Posted - 10/25/2005 : 5:19:34 PM
Hi Michele,
Try to replace the 2-4th lines to:
stat.worksheet$=Data1;
stat.mr.YData$=P; //dependent variable
stat.mr.xData1$=b; //indep. variable b
stat.mr.xData2$=c; //indep. variable c
Hope this works okay.
--Hideo Fujii
OriginLab
|
|
|
