[R-sig-ME] Non-Linear Mixed Effect Model - 4 vs. 3 parameter Sigmoid

[Timothy Lau]

`?nlmer` gives a quick example of a non-linear regression model using the
datasets::Orange dataset.

`nlmer(formula = circumference ~ SSlogis(input = age, Asym = Asym, xmid =
xmid, scal = scal) ~ Asym | Tree, start = c(Asym = 200, xmid = 725, scal =
350), nAGQ = 1, data = orange2)`

I wanted to switch the 3 parameter stats::SSlogis() for the 4 parameter
FlexParamCurve::SSposnegRichards()

```
m1_start_val <- getInitial(circumference ~ SSposnegRichards(x = age, Asym =
Asym, K = K, Infl = Infl, M = M, modno = 12, pn.options = "myoptions"),
data = Orange)
m1_parm4 <- nlmer(formula = circumference ~ SSposnegRichards(x = age, Asym
= Asym, K = K, Infl = Infl, M = M,, modno = 12, pn.options = "myoptions") ~
Asym | Tree,
      start = m1_start_val,
      nAGQ = 1,
      verbose = T,
      data = Orange)
```

But I keep getting errors, with the above code I get that "Error in
mkRespMod(fr, nlenv = nlenv, nlmod = nlmod) : is.matrix(gr <- attr(val,
"gradient")) is not TRUE"

Does anyone have some advice for a better way to get this to run?



Best,
Timothy


The will to win means nothing without the will to prepare.
Enthusiasm is common. Endurance is rare.
- J. Ikangaa; A. Duckworth;

	[[alternative HTML version deleted]]