smooth.construct.pcre.smooth.spec

a smooth specification object, see
  <code><a href="/link/smooth.construct?package=refund&version=0.1-1&to=mgcv" rd-options="mgcv" data-mini-rdoc="mgcv::smooth.construct">smooth.construct</a></code>

object

see <code><a href="/link/smooth.construct?package=refund&version=0.1-1&to=mgcv" rd-options="mgcv" data-mini-rdoc="mgcv::smooth.construct">smooth.construct</a></code>

data

knots

Sets up design matrix for functional random effects based
on the PC loadings of the covariance operator of the random
effect process. See
<code><a href="/link/smooth.construct.re.smooth.spec?package=refund&version=0.1-1&to=mgcv" rd-options="mgcv" data-mini-rdoc="mgcv::smooth.construct.re.smooth.spec">smooth.construct.re.smooth.spec</a></code> for
more details on <code>mgcv</code>-style smoother specification
and <code><a href="/link/pcre?package=refund&version=0.1-1" rd-options="" data-mini-rdoc="refund::pcre">pcre</a></code> for the corresponding
<code>pffr()</code>-formula wrapper.

Functions for regression with functional data.  The
        principal methods currently implemented are (1)
        functional-response/scalar-predictor regression by penalized
        OLS or GLS; and (2) functional principal component regression
        for scalar responses and functional predictors.

smooth.construct.pcre.smooth.spec: mgcv-style constructor for PC-basis functional random effects

Description

Usage

Arguments

Value