parallel_iter.Rd
This function performs a single iteration of the \(\beta\) method for A item's validation. It is designed
to be used in parallel computing environments to speed up the validation process of the \(\beta\) method.
The function is a utility function for validation
, and it should not be called independently by the user.
parallel_iter(
i,
Y,
P.alpha.Xi,
P.alpha,
pattern,
ri,
Ni,
Q.pattern.ini,
model,
criter,
search.method,
P_GDINA,
Q.beta,
L,
K,
alpha.P,
get.MLRlasso,
priority
)
Item number that need to be validated.
Observed data matrix for validation.
Individual posterior
Attribute prior weights.
The attribute mastery pattern matrix.
A vector that contains the numbers of examinees in each knowledge state who correctly answered item \(i\).
A vector that contains the total numbers of examinees in each knowledge state.
Initial pattern number for the model.
Model object used for fitting (e.g., GDINA).
Fit criterion ("AIC", "BIC", "CAIC", or "SABIC").
Search method for model selection ("beta", "ESA", "SSA", or "PAA").
Function to calculate probabilities for GDINA model.
Q-matrix for validation.
Number of latent pattern.
Number of attributes.
Individuals' marginal mastery probabilities matrix (Tu et al., 2022)
Function for Lasso regression with multiple linear regression.
Vector of priorities for PAA method search.
An object of class validation
is a list
containing the following components:
The previous fit index value after applying the selected search method.
The current fit index value after applying the selected search method.
The pattern that corresponds to the optimal model configuration for the current iteration.
The priority vector used in the PAA method, if applicable.