How to compare nonlinear regression coefficients across different data sets? I have looking for the answer a whole day. Thanks for your advice!
It is better to specify a larger model that estimates the difference in coefficients between the "data sets" (or groups) directly.
I agree with Jochen. But please also clarify what you mean by "nonlinear regression coefficients". Sometimes, people use "nonlinear" to describe models that are actually linear models (e.g., when they include polynomial terms). If you describe exactly what kind of model you're talking about, and indicate what software you use, someone may be able to give more precise advice. HTH.
Thank you very much, I have read all of your advice
The equation is R=αexp(βT), α and β are parameters, R is the dependent variable and T is the independent variable.
I have three datasets, and decide to fit the exponential model using these three datasets respectively, and then test if there are significant differences in parameters ( α and β ) for different datasets. And I try to use the R software. I found some authors used the "nlme" package to do this, but there was no details, and I have tried to looking for the help text of "nlme" , Unfortunately, it is difficult for me, I'm a newcomer of R language.
Reading suggestion: Nonlinear models in R, ISBN 0387096159
It explains several ways how to specify a nonlinear model over a design matrix.
- Badges
- Science topic