estimate: Function to estimate a choice model using cmdlr

In edsandorf/cmdlR: Choice Modeling in R

Function to estimate a choice model using cmdlr

Description

either maximize or minimize the log-likelihood function. If estimation is successful, the function calculates a high-precision numerical hessian and a standard and robust variance-covariance matrix.

Usage

estimate(ll, estim_env, model_options, control = NULL)

Arguments

ll	This is the 'raw' log-likelihood function passed to the estimation routine. It is important that the user takes into account whether the optimization routine is a minimizer (e.g. 'ucminf') or a maximizer (e.g. 'maxlik').
estim_env	An estimation environment or list of estimation environments returned from prepare
model_options	A list of model options. Note, this list is validated for a second time here to set some necessary defaults. See validate for details.
control	A list of control options that are passed to set_controls.

Details

NOTE: There are no checks on whether the log-likelihood function is correctly specified. If you see that your log-likelihood value is getting progressively worse from starting values, check your 'll' specification and that it corresponds to the chosen optimizer.

Value

A 'cmdlr' model object

Examples

## Not run: 
  # See /examples for how to use
  estimate(ll, estim_env, model_options)
  
  # or 
  estimate(ll, estim_env, model_options, control)

## End(Not run)

edsandorf/cmdlR documentation built on Jan. 17, 2024, 12:33 a.m.