I am starting an RSM d-optimal design, 5 numeric, 1 categoric variable, but I am a little confused. I don't know if it is strictly necessary to follow RSM protocol and start with a first degree equation as a first order model, and then a second degree and second order equation (if the response shows curvature) as the textbooks dictate. Is it possible to start RSM with a second order equation?
Dear Carlos
I think this recommendation is due to the fact that as we are starting a new experiment we don't know which model will adjust the studied phenomena and we have to check step by step. But it is clear that the first order model never works when you have more than 2 variables.
Regards!
NOBOSSE.
In RSM analysis the data has to be fitted to to a second order polynomial.
- Badges
- Science method
- Similar topics
- Methodology