Skip to contents

Generic for nlmixr2 estimation methods

Source: R/agq.R, R/bobyqa.R, R/emviControl.R, and 39 more
nlmixr2Est.Rd

Generic for nlmixr2 estimation methods

Usage

# S3 method for class 'agq'
nlmixr2Est(env, ...)

# S3 method for class 'bobyqa'
nlmixr2Est(env, ...)

# S3 method for class 'emvi'
nlmixr2Est(env, ...)

# S3 method for class 'fbvi'
nlmixr2Est(env, ...)

# S3 method for class 'fo'
nlmixr2Est(env, ...)

# S3 method for class 'foce'
nlmixr2Est(env, ...)

# S3 method for class 'focei'
nlmixr2Est(env, ...)

# S3 method for class 'output'
nlmixr2Est(env, ...)

# S3 method for class 'foceif'
nlmixr2Est(env, ...)

# S3 method for class 'focef'
nlmixr2Est(env, ...)

# S3 method for class 'focepf'
nlmixr2Est(env, ...)

# S3 method for class 'mfoceif'
nlmixr2Est(env, ...)

# S3 method for class 'mfocef'
nlmixr2Est(env, ...)

# S3 method for class 'mfocepf'
nlmixr2Est(env, ...)

# S3 method for class 'ifoceif'
nlmixr2Est(env, ...)

# S3 method for class 'ifocef'
nlmixr2Est(env, ...)

# S3 method for class 'ifocepf'
nlmixr2Est(env, ...)

# S3 method for class 'agqf'
nlmixr2Est(env, ...)

# S3 method for class 'magqf'
nlmixr2Est(env, ...)

# S3 method for class 'iagqf'
nlmixr2Est(env, ...)

# S3 method for class 'flaplace'
nlmixr2Est(env, ...)

# S3 method for class 'mflaplace'
nlmixr2Est(env, ...)

# S3 method for class 'iflaplace'
nlmixr2Est(env, ...)

# S3 method for class 'fagq'
nlmixr2Est(env, ...)

# S3 method for class 'mfagq'
nlmixr2Est(env, ...)

# S3 method for class 'ifagq'
nlmixr2Est(env, ...)

# S3 method for class 'focep'
nlmixr2Est(env, ...)

# S3 method for class 'foi'
nlmixr2Est(env, ...)

# S3 method for class 'iagq'
nlmixr2Est(env, ...)

# S3 method for class 'ifoce'
nlmixr2Est(env, ...)

# S3 method for class 'ifocei'
nlmixr2Est(env, ...)

# S3 method for class 'ifocep'
nlmixr2Est(env, ...)

# S3 method for class 'ilaplace'
nlmixr2Est(env, ...)

# S3 method for class 'imp'
nlmixr2Est(env, ...)

# S3 method for class 'impmap'
nlmixr2Est(env, ...)

# S3 method for class 'laplace'
nlmixr2Est(env, ...)

# S3 method for class 'lbfgsb3c'
nlmixr2Est(env, ...)

# S3 method for class 'magq'
nlmixr2Est(env, ...)

# S3 method for class 'mfoce'
nlmixr2Est(env, ...)

# S3 method for class 'mfocei'
nlmixr2Est(env, ...)

# S3 method for class 'mfocep'
nlmixr2Est(env, ...)

# S3 method for class 'mlaplace'
nlmixr2Est(env, ...)

# S3 method for class 'n1qn1'
nlmixr2Est(env, ...)

# S3 method for class 'newuoa'
nlmixr2Est(env, ...)

# S3 method for class 'nlm'
nlmixr2Est(env, ...)

# S3 method for class 'nlme'
nlmixr2Est(env, ...)

# S3 method for class 'nlminb'
nlmixr2Est(env, ...)

nlmixr2Est(env, ...)

# Default S3 method
nlmixr2Est(env, ...)

# S3 method for class 'nls'
nlmixr2Est(env, ...)

# S3 method for class 'npag'
nlmixr2Est(env, ...)

# S3 method for class 'mnpag'
nlmixr2Est(env, ...)

# S3 method for class 'inpag'
nlmixr2Est(env, ...)

# S3 method for class 'npb'
nlmixr2Est(env, ...)

# S3 method for class 'mnpb'
nlmixr2Est(env, ...)

# S3 method for class 'inpb'
nlmixr2Est(env, ...)

# S3 method for class 'optim'
nlmixr2Est(env, ...)

# S3 method for class 'neldermead'
nlmixr2Est(env, ...)

# S3 method for class 'bfgs'
nlmixr2Est(env, ...)

# S3 method for class 'cg'
nlmixr2Est(env, ...)

# S3 method for class 'lbfgsb'
nlmixr2Est(env, ...)

# S3 method for class 'sann'
nlmixr2Est(env, ...)

# S3 method for class 'brent'
nlmixr2Est(env, ...)

# S3 method for class 'posthoc'
nlmixr2Est(env, ...)

# S3 method for class 'qrpem'
nlmixr2Est(env, ...)

# S3 method for class 'rxSolve'
nlmixr2Est(env, ...)

# S3 method for class 'simulate'
nlmixr2Est(env, ...)

# S3 method for class 'simulation'
nlmixr2Est(env, ...)

# S3 method for class 'predict'
nlmixr2Est(env, ...)

# S3 method for class 'saem'
nlmixr2Est(env, ...)

# S3 method for class 'trust'
nlmixr2Est(env, ...)

# S3 method for class 'uobyqa'
nlmixr2Est(env, ...)

# S3 method for class 'vae'
nlmixr2Est(env, ...)

Arguments

env

Environment for the nlmixr2 estimation routines.

This needs to have:

  • rxode2 ui object in $ui

  • data to fit in the estimation routine in $data

  • control for the estimation routine's control options in $ui

...

Other arguments provided to nlmixr2Est() provided for flexibility but not currently used inside nlmixr

Value

nlmixr2 fit object

Details

This is a S3 generic that allows others to use the nlmixr2 environment to do their own estimation routines

A prior distribution given in the ini({}) block is never silently ignored. Before dispatching, nlmixr2Est() refuses any prior the method cannot use, so a method that does not handle priors gets that for free and a model carrying one fails with an explanation instead of being fit to something other than what it says.

A method says what it supports with an attribute on itself:

attr(nlmixr2Est.myMethod, "nlmixr2Priors") <- "general"

  • "none", which is also what an absent attribute means – the method cannot use priors at all

  • "theta" – priors on population parameters only (dnorm(), dcauchy(), stdNormal() and the multiNormal() family among thetas); anything referencing an omega element is refused. No FOCEi-family method uses this level – they declare "general".

  • "general" – everything the shared kernel's "general" method covers: the above, plus a normal prior directly on an omega element and a textbook inverse-Wishart on an omega block. This is what est="focei" and the rest of its family declare.

  • "nwpri" – normal priors and degrees of freedom on an omega block (invWishart(4)), evaluated the way a NONMEM NWPRI model works; a normal prior on the omega values themselves is refused

  • "tnpri" – normal priors including directly on omega elements (Monolix's/NONMEM's own-estimation joint-normal assumption); dcauchy() and invWishart() are refused

  • "all" – the method handles everything, so nothing is checked

A repeated omega block (same()) ties its copies' variances to its master's. A model with one is refused by any method that has not said it keeps that tie, so a method that would estimate the copies apart fails with an explanation instead of dropping the tie. A method says so with

attr(nlmixr2Est.myMethod, "sameBlock") <- TRUE

when it estimates the copies as one, and the refusal then names it among the methods that keep the tie; or with

attr(nlmixr2Est.myMethod, "sameBlock") <- "fixed"

when it does not estimate omega at all (it takes the model's omega as given), so it is accepted but not named. attr(nlmixr2Est.myMethod, "iovSame") <- TRUE, which lets a method use the repeated-block expansion of inter-occasion variability (iovMethod = "omega"), says the same as "sameBlock" TRUE.

Author

Matthew Fidler