I did the experiment based on response surface method (central composite) in design expert software. At the end of the experiment, I had a quadratic regression model that it can used for predicting the response. also using optimizing section of design expert software, can be used for predicting. but there is a problem that I can't understand it: predicted value of response with equation that manually was done (it is 9) is not equal to predicting optimizing section of software (it is 12).why they are different? Also I check some articles and this issue exist in all of them.
thanks for guides.
Without details, it is hard to say but my guess is the prediction model is giving you the average response and the optimizing section the estimated optimal response. Please check your software manual. Best wishes, David Booth
David is right, there are no details. Please provide R2, adjusted R2 and Pred R2 values. Remember, that the Pred R2 value and its difference from adjusted R2 (which should be in the region of 0.2) will say a lot about the suitability of your model for predicting responses for an object that was not used for model estimation. I don't also know if you considered the desirability function during optimization. Check this too and like Dave said consult the software manual.
thanks so much for your answers.
R2, adjusted R2 and Pred R2 values are 0.98, 0.97 and 0.93, respectively. so the Pred R2 difference from adjusted R2 is less than 0.2 and it's OK. about the desirability of software predicting (not predicting of equation). it's equal 1. so it's OK too. but I don't know what's your mean about the software manual. please guide me if possible.
thanks very much.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression