I think it can simply be solved by linearizing the equations, e.g. log(phi) = log(beta1) + beta2*covariate1 instead of phi = beta1 * exp(beta2*covariate1) and including this accordingly in the definition of the fitting function.
nlmefit and covariate building
1 view (last 30 days)
Show older comments
Hi, I try to use nlmefit on PK data. It seems like the inclusion of covariates is limited to linear relationships, e.g. phi = beta1 + covariate1 * beta2 etc ... Is it possible to include relationships like exponential or power functions, e.g. phi = beta1 * covariate1^beta2 or phi = beta1 * e^(beta2*covariate1)? Thanks in advance
0 Comments
Answers (1)
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!