I’ve got a data set and I want to calculate R2 for linear regression equation from another study.
For example, I have derived an equation from my data (with R2) and I want to test how other equations perform on my data (and thus calculate R2 for them). Then, I want to compare R2 from my data set with R2 from derivation studies.
Do you have any software for this? Any common statistical software could cope with this task (e.g. SPSS or SAS)? Maybe you have any tutorials on YouTube for this?
For example here is a study where researchers calculated R2 for their cohort based on previously derived regression equations:
Article Normative cardiopulmonary exercise data for endurance athlet...
.
When you say you want to compare R2 values, do you mean you want a formal statistical comparison of them, or do you just want to apply the so-called eyeball test? ;-) And how many other sets of coefficients do you want to use?
If all you want to use the eyeball test with one other set of coefficients, something quick and dirty like this SPSS code should do the trick.
* Modify path on the next line as needed.
GET FILE = "C:/SPSSdata/survey_sample.sav".
* Regress educ on age, paeduc and maeduc.
* Save the fitted values.
REGRESSION
/STATISTICS COEFF OUTS CI(95) R ANOVA
/DEPENDENT educ
/METHOD=ENTER age paeduc maeduc
/SAVE PRED(yhat1).
* Now use another set of coefficients to generate yhat2.
COMPUTE yhat2 = 10 + .015*age + .15*paeduc + .11*maeduc.
REGRESSION
/STATISTICS R
/DEPENDENT educ
/METHOD=ENTER yhat1.
REGRESSION
/STATISTICS R
/DEPENDENT educ
/METHOD=ENTER yhat2.
I am taking advantage of the fact that the multiple correlation (R) for a model = the Pearson correlation (r) between Y and the fitted value of Y.
The COMPUTE command for yhat2 could be automated (using OMS etc.), but if you have just one set of coefficients to plug in, it may not be worth the trouble of doing that.
HTH.
Bruce Weaver
I just want to apply eyeball test. I do not want to use any other statistics to compare calculated R2.
I want to use the same coefficients as in the originally derived regression equations.
In other words, I want to validate the existing regression equations and test how they perform on my sample (please see the Table 3. from the proposed paper).
Article Normative cardiopulmonary exercise data for endurance athlet...
Murio Joel
I do not analyze data in R programming language.
I just want to calculate R2 (coefficient of determination) for existing regression equations.
I want to do this with the usage of other software, perhaps SPSS or SAS.
Hello Pavel Makovsky. In that case, modify the code I posted earlier. With your dataset open and active:
REGRESSION
/STATISTICS COEFF OUTS CI(95) R ANOVA
/DEPENDENT YourDV
/METHOD=ENTER your explanatory variables
/SAVE PRED(yhat1).
* Now use another set of coefficients to generate yhat2.
COMPUTE yhat2 = use the other regression coefficients here but with the variables in your dataset.
REGRESSION
/STATISTICS R
/DEPENDENT YourDV
/METHOD=ENTER yhat1.
REGRESSION
/STATISTICS R
/DEPENDENT YourDV
/METHOD=ENTER yhat2.
Oluwaseyi Ayorinde Mohammed
Bruce Weaver
Murio Joel
Thank you for you help!
Shriniwas Bharati Could you write the procedure with all steps for MiniTab?
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression