How to get all R squared values in Stepwise regression?

Hello,

I have a dataset and I want to do some sensitivity analysis to rank my independent variables based on their effect on the output. I want to use Stepwise regression. However, it just shows the final and overall R squared value (for the final model) when it finishes its job. Is there any way to force Matlab to print R squared values in each step, while it's adding one new term to the model?

So, in each step, I will be able to measure the importance of each term.

Thank you,

Answer 1

[the cyclist]

If you are using stepwisefit(), look at the "history" output, which has the RMSE value at each step. You can calculate R^2 from the RMSE.

If you are using stepwise(), which is the interactive version of stepwise regression, you can see the value of R^2 at each step.

[Ehsan]

Hi,

Thanks for your answer. Would you please explain more? I found this, written by Tom Lane:

"R^2 = 1 - SSE/SST = 1 - DFE*RMSE^2/SST Here SSE is the error sum of squares, SST is the total sum of squares, and DFE is the degrees of freedom for error." http://www.mathworks.com/matlabcentral/answers/36351-relationship-between-rmse-and-r-2

How can I get all these values when I am using stepwisefit()? Would you please give me an example?

Thank you.