my.samp.cov2d: compute sample covariance for functional data

In dan410/sseigfun: SS-ANOVA covariance function estimation and eigenfunction estimation

Description

compute sample covariance for functional data

Usage

my.samp.cov2d(time, x, subject, wt, marginal.knots, n.marginal.knots,
  centered = TRUE, noDiag = TRUE)

Arguments

time	vector of points on the 'time' axis
x	vector of functional observations
subject	vector of integer IDs for each curve
wt	vector of weights. These weights are not used to calculate anything in this function. The wt argument here is necessary for 'bookeeping' purposes for covariance function estimation methods that utilize these weights.
marginal.knots	vector of knot locations on the 'time' (i.e. marginal) axis.
n.marginal.knots	integer specifying the number of knot locations to use on the marginal domain
centered	logical. If FALSE a smoothing spline fit will be computed to estimate the mean functions, and this mean funciton will be subtracted from each curve.
noDiag	logical. If TRUE the diagonal elements of th estimated covariance matrix will be removed.

Details

These weights argument wt is not used to calculate anything in this function. The wt argument here is necessary for 'bookeeping' purposes for covariance function estimation methods that utilize these weights.

Value

A data frame with the sample covariances. The knot locations are included at the end of the data frame.

dan410/sseigfun documentation built on May 14, 2019, 3:34 p.m.