param_est_linear: Imputation Parameter Estimation for Linear Regression

In justin-petrovich/sparsefreg: Scalar-on-Function Regression with a Highly Irregularly-Sampled Functional Covariate

Description

Given the data, this function estimates the imputation parameters. This function is a wrapper for fcr if cond.y is set to TRUE, and face.sparse if cond.y is set to FALSE.

Usage

param_est_linear(
  dat,
  y,
  workGrid,
  M,
  cond.y = TRUE,
  use_fcr = TRUE,
  fcr.args = list(use_bam = T, niter = 1),
  k = -1,
  face.args = list(knots = 12, lower = -3, pve = 0.95),
  FPCA.args = NULL
)

Arguments

dat	A data frame with n rows (where N is the number of subjects, each with m_i observations, so that ∑_{i=1}^N m_i = n) expected to have either 3 or 4. If cond.y is TRUE, should include 4 columns, with variables 'X','y','subj', and 'argvals'. If cond.y is FALSE, only 3 columns are needed (no 'y' variable is used).
workGrid	A length M vector of the unique desired grid points on which to evaluate the function.
cond.y	A logical indicator to determine whether imputation will be done conditional on Y.
fcr.args	A list of arguments to be passed to the underlying function fcr.
k	Dimension of the smooth terms used in fcr. Only needs to be specified if cond.y is TRUE.
face.args	A list of arguments to be passed to the underlying function face.sparse.
nPhi	An integer value, indicating the number of random effects to include in the model. This is passed to fcr and is only used when cond.y is TRUE. See fcr for more details.

justin-petrovich/sparsefreg documentation built on Aug. 20, 2020, 9:04 p.m.