RapidMiner 9.7 is Now Available

Lots of amazing new improvements including true version control! Learn more about what's new here.

CLICK HERE TO DOWNLOAD

"How to set up multiple regression analysis?"

DonTDonT Member Posts: 1 Contributor I
edited June 2019 in Help
Hello everyone.  I'm very new to RM, and got here when I was looking for tools to help me run statistical inference tests on some data.  I've searched through previous posts on this topic but haven't quite found an answer.

Assume I have one "dependent" variable -- some measurement of interest, such as "age when death occurs".  Call this variable Y.
Assume I have  3 "independent" variables, that could be measures of "unhealthy" factors: X1=number of pounds overweight; X2=cholesterol level; X3=triglyceride level.  Assume I have 100 (deceased) people with measurements on all this variables.

I want to run a multiple regression analysis http://en.wikipedia.org/wiki/Regression_analysis to see how well the combination of 3 independent variables "predicts" the dependent variable, and end up with a value for F (and its significance level), a value for R squared, and the regression coefficients.

I see the Linear Regression and Vector Linear Regression operators.  I also see the ANOVA operator, and the ANOVA Matrix and Grouped ANOVA operators.  (I don't know how to obtain a Performance Vector, which is required for the ANOVA operator.)  But I don't see how to get what I need using some combination of these and other operators.  Is RM not intended for this kind of inferential significance testing?

Different but related question:  Assume I have one dependent variable Y, 2 independent variables A & B, and a 2x2 experimental design: Can RM calculate the 3 ANOVA F values to indicate whether variable A, and/or B, and or their AxB interaction, have statistically significant effects on Y?

Thanks for any and all tips!
 
Tagged:
Sign In or Register to comment.