Dear all,
Can anyone tell me how to fit nonlinear mixed models by the first-order (FO) and first-order conditional expectation (FOCE) methods ? I only know that a Package called nlme can be used to fit nonlinear mixed models by nlme(model, data, fixed, random, groups, start, correlation, weights, subset, method, na.action, naPattern, control, verbose) in R software.
But I don't know which method (FO or FOCE) was used in nlme exactly. Both FO and FOCE methods linearize a nonlinear mixed model through a first-order Taylor series expansion.They differ on how random parameter vector bi is predicted and how subsequent SS predictions are generated. I got that Nonlinear mixed models can be fitted by the SAS macro NLINMIX by incorporating random parameters into these two models.But how to do it in R software? And I upload an article about the question. Can you help me? Thank you very much!
HoPui
I am not familiar with the FOCE methods, but at first sight this seems to be feasible through the Phxnlme package, have you tried it?
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Software
- R