re_evaluate_model: Re-evaluate the model

In dRiftDM: Estimating (Time-Dependent) Drift Diffusion Models

Re-evaluate the model

Description

Updates the PDFs of a model. If obs_data are set to the model, the log-likelihood is also updated.

Usage

re_evaluate_model(drift_dm_obj, eval_model = TRUE)

Arguments

drift_dm_obj	an object of type drift_dm
eval_model	logical, indicating if the model should be evaluated or not. If FALSE, PDFs and the log-likelihood value are deleted from the model. Default is True.

Details

More in-depth information about the mathematical details for deriving the PDFs can be found in \insertCiteRichteretal.2023;textualdRiftDM

Value

Returns the passed drift_dm_obj object, after (re-)calculating the PDFs and (if observed data is set) the log-likelihood.

• the PDFs an be addressed via drift_dm_obj$pdfs

• the log-likelihood can be addressed via drift_dm_obj$log_like_val

Note that if re_evaluate model is called before observed data was set, the function silently updates the pdfs, but not log_like_val.

See Also

drift_dm()

Examples

# choose a pre-built model (e.g., the Ratcliff model)
# and set the discretization as needed
my_model <- ratcliff_dm(t_max = 1.5, dx = .005, dt = .005)

# then calculate the model's predicted PDF
my_model <- re_evaluate_model(my_model)
str(my_model$pdfs) # show the structure of the attached pdfs

# if you want the log_likelihood, make sure some data is attached to the
# model (see also the documentation of obs_data())
obs_data(my_model) <- ratcliff_synth_data # this data set comes with dRiftDM
my_model <- re_evaluate_model(my_model)
str(my_model$pdfs)
print(my_model$log_like_val)

dRiftDM documentation built on April 3, 2025, 7:48 p.m.