Fit XSL model using differential evolution

Fits a model to provided data using the Differential Evolution optimization algorithm. By default it optimizes the model parameters to minimize the sum of squared errors (SSE) between the model's predictions and human accuracy data; with control = xslControl(objective = "binomial") (or "multinomial") it instead maximizes the likelihood of the human test responses (see xslControl()).

Usage

xsl_fit(
  model,
  data,
  lower,
  upper,
  by_data = FALSE,
  control = xslControl(),
  deoptim_control = DEoptim::DEoptim.control(reltol = 0.001, NP = 100, itermax = 100),
  free = NULL,
  n_restarts = 1,
  seed = NULL
)

Arguments

model: An object of class xslMod.
data: An object (or list of objects) of class xslData.
lower: Numeric vector of lower bounds for the model's parameters.
upper: Numeric vector of upper bounds for the model's parameters.
by_data: Logical indicating whether to fit to each entry in data separately.
control: Control parameters passed to xsl_run().
deoptim_control: Control parameters passed to DEoptim().
free: Optional names of the parameters that lower/upper bound, in order (see update_params()); by default, the model's leading parameters. With a likelihood objective, "lapse" fits the control's lapse rate along with the model's parameters (e.g. bounds 0 to 1).
n_restarts: Number of independent DEoptim runs per fit; the best (lowest SSE) is kept. Some models' objectives have poor local basins that a single run lands in a good fraction of the time (e.g. uncfam_predictive()), so a few restarts make fits far more reliable.
seed: Optional random seed, set once before fitting, so DEoptim's random initial populations (and a stochastic model's simulations) are reproducible.

Value

A list with one element per fit (one in total, or one per dataset with by_data = TRUE), each the DEoptim result of the best run, which includes the best parameters found (optim$bestmem, named) and their SSE (optim$bestval; the negative log-likelihood with a likelihood objective). With n_restarts > 1, restarts holds every run's best value. With a likelihood objective, each fit also has loglik, n_obs (the number of human responses), k (the number of fitted parameters, including a fitted lapse), and aic and bic.

Examples

fit <- xsl_fit(decay(C = .98), xsl_datasets[1:2], lower = .8, upper = 1,
               n_restarts = 2, seed = 1,
               deoptim_control = DEoptim::DEoptim.control(NP = 10, itermax = 5, trace = FALSE))
fit[[1]]$optim$bestmem
#>         C 
#> 0.9269144 
fit[[1]]$restarts
#> [1] 0.2686 0.2686

# fit by binomial likelihood, with the lapse rate as a free parameter
fit <- xsl_fit(decay(C = .98), xsl_datasets[1:2], lower = c(.8, 0), upper = c(1, .5),
               free = c("C", "lapse"), control = xslControl(objective = "binomial"),
               seed = 1,
               deoptim_control = DEoptim::DEoptim.control(NP = 10, itermax = 5, trace = FALSE))
#> Warning: For many problems it is best to set 'NP' (in 'control') to be at least ten times the length of the parameter vector. 
fit[[1]]$optim$bestmem
#>         C     lapse 
#> 0.8920193 0.2176450 
fit[[1]]$aic
#> [1] 189.0766