An internal function that manages the block-wise GETS search within the
isat() type functions.
ISblocksFun(j, i, ISmatrices, ISblocks, mX, parallel.options, y, userEstArg,
t.pval, wald.pval, do.pet, arLjungB, archLjungB, normality.JarqueB,
user.diagnostics, gofFunArg, gof.method, mxkeep, include.gum, include.1cut,
include.empty, max.paths, turbo, tol, LAPACK, max.regs, print.searchinfo, y.n)Returns a character vector of the names of the retained variables or NULL.
Index of the current block (within the ith indicator matrix) to be evaluated.
Index of the indicator matrix from ISmatrices being processed (e.g., IIS, SIS, TIS).
A named list of indicator matrices generated by create.ISmatrices.
A list of block index sets corresponding to columns of each indicator matrix, specifying how indicators are grouped for block-wise GETS.
Matrix of base regressors to which indicator variables are appended.
Parallel execution control; if NULL, no parallelism is applied.
Numeric vector or time-series; the dependent variable in the regression model.
A list containing the user-defined estimation function and its arguments,
passed to getsFun.
Numeric value between 0 and 1; the significance level used for coefficient t-tests during GETS.
Numeric value; significance level for Wald tests in GETS selection.
Logical; if TRUE, a Parsimonious Encompassing Test (PET) is
conducted after each deletion.
A two-element list or vector specifying the lag and p-value for the Ljung-Box test on residual autocorrelation.
A two-element list or vector specifying the lag and p-value for the Ljung-Box test on squared residuals (ARCH test).
Numeric value or NULL; p-value threshold for the Jarque-Bera
test for normality.
A list specifying a user-defined diagnostics function and its significance levels.
List specifying the name and parameters of the goodness-of-fit function.
Character string; either "min" or "max" indicating how to
choose the best model based on the GOF criterion.
Indices of variables in mXis to be forcibly retained during the
GETS procedure.
Logical; if TRUE, includes the General Unrestricted Model in the
set of terminal models.
Logical; if TRUE, includes the 1-cut model in the set of terminal
models.
Logical; if TRUE, includes the empty model among terminal models
if it passes diagnostics.
Maximum number of paths to explore during GETS. If NULL,
no limit is imposed.
Logical; avoids redundant path evaluations in GETS for efficiency.
Numeric tolerance for detecting collinearity in regressors during model estimation.
Logical; if TRUE, LAPACK routines are used for matrix decomposition
in collinearity checks.
Maximum number of regressions allowed in each deletion path.
Logical; if TRUE, prints messages about the GETS search progress
and block structure.
Integer; the number of observations in the dependent variable y.
See also isat.