cckm: Fit Mixed Inputs Kriging Model

In dominikkirchhoff/CCKriging: Kriging for Mixed Continuous and Categorical Input Variables

Description

Given mixed continuous and categorical design points x and a continuous output y, fit an extended Kriging model. There a several methods available for handling the categorical inputs. The optimal parameter set is estimated via Maximum Likelihood Estimation.

Usage

cckm(x, y, cat.type, cont.type = "matern", nu = 2.5, pars.init = NULL,
  info = TRUE)

Arguments

x	[data.frame] Mixed continuous and categorical input points.
y	[data.frame] Continuous output variable.
cat.type	[character(1)] Which method should be used for the categorical inputs? Possible choices: “EC”: Exchangeable Correlation “MC”: Multiplicative Correlation “UC”: Hypersphere Decomposition-Based Unrestrictive Correlation “TMC”: Toeplitz Matrix Multiplication-Based Correlation
cont.type	[character(1)] Which correlation function should be used for the continuous inputs? Possible choices: “matern” : Matern correlation function
nu	[numeric(1)] Parameter nu for the Matern correlation function. Defaults to 2.5.
pars.init	[numeric] Optional vector of initial paramters for the optimization. Defaults to NULL, which means the best of 100 randomly generated parameter settings is used as the initial setting.
info	[logical(1)] Should extra information be printed to the console? Default is TRUE.

dominikkirchhoff/CCKriging documentation built on May 19, 2019, 4:05 p.m.