Use emperical Bayes estimates to simulate re-using ETAs

Simulation reusing ETA values from estimation run or otherwise specified ETA values. For observed subjects, this is refered to as emperical Bayes estimates (EBE). The .phi file from the estimation run must be found next to the .lst file from the estimation.This means that ID values in the (simulation) input data must be ID values that were used in the estimation too. Runs an $ESTIMATION MAXEVAL=0 but pulls in ETAs for the ID's found in data. No $SIMULATION step is run which may affect how for instance residual variability is simulated, if at all. You can also specify a different .phi file which can be a simulation result.

Usage

NMsim_EBE(file.sim, file.mod, data.sim, file.phi, return.text = FALSE)

Arguments

file.sim

The path to the control stream to be edited. This function overwrites the contents of the file pointed to by file.sim.

file.mod

Path to the path to the original input control stream provided as `file.mod` to `NMsim()`.

data.sim

See ?NMsim.

file.phi

A phi file to take the known subjects from. The default is to replace the filename extension on file.mod with .phi. A different .phi file would be used if you want to reuse subjects simulated in a previous simulation.

return.text

If TRUE, just the text will be returned, and resulting control stream is not written to file.

Value

Path to simulation control stream

See also

simPopEtas