G.cd: Family of continuous/discrete instrumental function

View source: R/BoundingCovariateEffects.R

G.cdR Documentation

Family of continuous/discrete instrumental function

Description

The function normalizes the continuous covariates to lie in the unit interval and then evaluates the subvector of continuous covariates on the specified family of instrumental function. For the discrete elements, indicator functions are used for each level.

Usage

G.cd(
  x,
  g.idx,
  data,
  n.if.per.cov,
  idxs.c,
  G.c,
  norm.func,
  discrete.covariate.levels = NULL,
  cov.ranges = NULL,
  norm.cov.out = NULL,
  degree = 3
)

Arguments

x

The vector of covariates at which to evaluate the B-splines

g.idx

The index of the instrumental function.

data

Data frame containing the data.

n.if.per.cov

Number of instrumental functions per continuous covariate.

idxs.c

Vector of indices of the continuous elements in the vector of covariates.

G.c

Family of instrumental functions to use for the subvector of continuous covariates.

norm.func

Function to be used to normalize the covariates.

discrete.covariate.levels

Matrix containing as rows all possible 'combined' levels of the discrete covariates. Default is discrete.covariate.levels = NULL.

cov.ranges

Matrix containing as its rows the lower and upper bounds for each continuous covariate. Default is cov.ranges = NULL.

norm.cov.out

Output of a preliminary call to a covariate normalization function (defined above). This is used to speed up computations. Note that this argument should only apply to continuous covariates!! Default is norm.cov.out = NULL.

degree

Degree of the spline functions to be used as instrumental functions for the continuous covariates (if applicable). Default is degree = 3.


depCensoring documentation built on April 4, 2025, 1:52 a.m.