ivgets: General-to-specific modeling for 2SLS models
In ivgets: General to Specific Modeling and Indicator Saturation in 2SLS Models

ivgets

R Documentation

General-to-specific modeling for 2SLS models

Description

General-to-specific modeling for 2SLS models

Usage

ivgets(
  formula,
  data,
  gum.result = NULL,
  t.pval = 0.05,
  wald.pval = t.pval,
  do.pet = TRUE,
  ar.LjungB = NULL,
  arch.LjungB = NULL,
  normality.JarqueB = NULL,
  include.gum = FALSE,
  include.1cut = FALSE,
  include.empty = FALSE,
  max.paths = NULL,
  turbo = FALSE,
  tol = 1e-07,
  max.regs = NULL,
  print.searchinfo = TRUE,
  alarm = FALSE,
  keep_exog = NULL,
  overid = NULL,
  weak = NULL
)

Arguments

`formula`	A formula in the format `y ~ x1 + x2 \| z1 + z2`.
`data`	A data frame with all necessary variables y, x, and z.
`gum.result`	a `list` with the estimation results of the General Unrestricted Model (GUM), or `NULL` (default). If the estimation results of the GUM are already available, then re-estimation of the GUM is skipped if the estimation results are provided via this argument
`t.pval`	`numeric` value between 0 and 1. The significance level used for the two-sided regressor significance t-tests
`wald.pval`	`numeric` value between 0 and 1. The significance level used for the Parsimonious Encompassing Tests (PETs)
`do.pet`	`logical`. If `TRUE` (default), then a Parsimonious Encompassing Test (PET) against the GUM is undertaken at each regressor removal for the joint significance of all the deleted regressors along the current path. If `FALSE`, then a PET is not undertaken at each regressor removal
`ar.LjungB`	a two element `vector` or `NULL` (default). In the former case, the first element contains the AR-order, the second element the significance level. If `NULL`, then a test for autocorrelation is not conducted
`arch.LjungB`	a two element `vector` or `NULL` (default). In the former case, the first element contains the ARCH-order, the second element the significance level. If `NULL`, then a test for ARCH is not conducted
`normality.JarqueB`	`NULL` or a `numeric` value between 0 and 1. In the latter case, a test for non-normality is conducted using a significance level equal to `normality.JarqueB`. If `NULL`, then no test for non-normality is conducted
`include.gum`	`logical`. If `TRUE`, then the GUM (i.e. the starting model) is included among the terminal models. If `FALSE` (default), then the GUM is not included
`include.1cut`	`logical`. If `TRUE`, then the 1-cut model is added to the list of terminal models. If `FALSE` (default), then the 1-cut is not added, unless it is a terminal model in one of the paths
`include.empty`	`logical`. If `TRUE`, then the empty model is added to the list of terminal models. If `FALSE` (default), then the empty model is not added, unless it is a terminal model in one of the paths
`max.paths`	`NULL` (default) or an integer equal to or greater than 0. If `NULL`, then there is no limit to the number of paths. If an integer, for example 1, then this integer constitutes the maximum number of paths searched
`turbo`	`logical`. If `TRUE`, then (parts of) paths are not searched twice (or more) unnecessarily, thus yielding a significant potential for speed-gain. However, the checking of whether the search has arrived at a point it has already been comes with a slight computational overhead. Accordingly, if `turbo=TRUE`, then the total search time might in fact be higher than if `turbo=FALSE`. This happens if estimation is very fast, say, less than quarter of a second. Hence the default is `FALSE`
`tol`	numeric value (`default = 1e-07`). The tolerance for detecting linear dependencies in the columns of the variance-covariance matrix when computing the Wald-statistic used in the Parsimonious Encompassing Tests (PETs), see the `qr.solve` function
`max.regs`	`integer`. The maximum number of regressions along a deletion path. Do not alter unless you know what you are doing!
`print.searchinfo`	`logical`. If `TRUE` (default), then a print is returned whenever simiplification along a new path is started
`alarm`	`logical`. If `TRUE`, then a sound or beep is emitted (in order to alert the user) when the model selection ends
`keep_exog`	A numeric vector of indices or a character vector of names corresponding to the exogenous regressors in the `data` that should not be selected over. Default `NULL` means that selection is over all exogenous regressors. If an intercept has been specified in the `formula` but is not already included in the `data`, then it can be kept by either including the index `0` or the character `"Intercept"`, respectively, as an element in `keep_exog`.
`overid`	`NULL` if no Sargan test of overidentifying restrictions should be used as a diagnostic check for model selection or a numeric value between 0 and 1. In the latter case, the test is conducted using this value as the significance level.
`weak`	`NULL` if no weak instrument F-test on the first stage should be used as a diagnostic check for model selection or a numeric value between 0 and 1. In the latter case, the test is conducted using this value as the significance level.

Value

Returns a list of class "ivgets" with three named elements. $selection stores the selection results from getsFun (including paths, terminal models, and best specification). $final stores the ivreg model object of the best specification or NULL if the GUM does not pass all diagnostics. $keep stores the names of the regressors that were not selected over, including the endogenous regressors, which are always kept.