frml.additive.smooth

integer, number for the variable to be regressed

target.ind

integer(s), number(s) for the variable(s) on which we regress

pred.inds

name of variables used to create formula, default is "x"

var.str


Creates a formula for <a rd-options="" href="/link/gam?package=kpcalg&version=1.0.1" data-mini-rdoc="kpcalg::gam">gam</a> to be used in <a rd-options="" href="/link/regrXonS?package=kpcalg&version=1.0.1" data-mini-rdoc="kpcalg::regrXonS">regrXonS</a>. For data $X=(X_1,...X_n,X_{n+1},...,X_m)$, variable to be regressed $X_{i}$, i=1...n and variables to regress on $S={X_{n+1},...,X_m}$ creates formula $X_i ~ s(X_{n+1}) + ... + s(X_m)$.


Kernel PC (kPC) algorithm for causal structure learning and causal inference using graphical models. kPC is a version of PC algorithm that uses kernel based independence criteria in order to be able to deal with non-linear relationships and non-Gaussian noise.

Petras Verbyla

kpcalg

Kernel PC Algorithm for Causal Structure Detection

frml.additive.smooth function


Creates a formula for <a rd-options='' href='gam'>gam</a> to be used in <a rd-options='' href='regrXonS'>regrXonS</a>. For data $X=(X_1,...X_n,X_{n+1},...,X_m)$, variable to be regressed $X_{i}$, i=1...n and variables to regress on $S={X_{n+1},...,X_m}$ creates formula $X_i ~ s(X_{n+1}) + ... + s(X_m)$.


frml.additive.smooth: Formula for GAM without crossterms

Description

Usage

Arguments

Value

See Also