eml_twochoiceRL: Expectation-Maximization with Laplace approximation (EML) of...

In psuthaharan/twochoiceRL: Modeling Behavior for Two-Choice Decisions

Description

This function runs the EML technique to recover behavior in a two-choice decision task.

Usage

eml_twochoiceRL(
  data = NULL,
  param = list(init_x1, init_x2),
  prior_mean = c(m1, m2),
  prior_sd = c(s1, s2),
  fn = "eml.objectiveFunction",
  opt = "TRM",
  radius = c(tr_rad[1], tr_rad[2]),
  nRes = nRes,
  iter = iter,
  eml_data = eml_data,
  progress_bar = progress_bar
)

Arguments

data	simulated two-choice task data
param	randomly generated initial parameters.
prior_mean	mean priors for x1 and x2, respectively. Defaults to mean 0 for both parameters.
prior_sd	standard deviation priors for x1 and x2, respectively. Defaults to 5 for both parameters.
fn	objective function being minimized
opt	optimization algorithm used. Defaults to trust-region method (trm)
radius	starting and maximum allowed trust region radius. Defaults to 1 and 5.
nRes	number of restarts. Defaults to 5.
iter	iteration of algorithm
eml_data	data to be passed for succeeding iterations
progress_bar	track completion time of estimation. Defaults to TRUE.

Value

A list containing the sum log-likelihood, eml estimates of parameter 1 per subject, eml estimates of parameter 2 per subject, laplace values of parameter 1 per subject, laplace values of parameter 2 per subject.

psuthaharan/twochoiceRL documentation built on Dec. 22, 2021, 9:56 a.m.