mlegp-naming-functions: mlegp naming functions

In mlegp: Maximum Likelihood Estimates of Gaussian Processes

mlegp naming functions

Description

Functions that set the design matrix parameter names in a Gaussian process or Gaussian process list or the names of the Gaussian processes in a Gaussian process list

Usage

setParams(x, s)
setGPNames(x, s)

Arguments

x	an object of type gp or gp.list
s	a vector of parameter names or Gaussian process names to set

Details

setParams sets the parameter names of a Gaussian process (gp) object or of all Gaussian processes in a Gaussian process list (gp.list) object. setGPNames sets the names of the Gaussian processes in an object of type gp.list

Value

the object x with parameter or Gaussian process names set

Note

Parameter and Gaussian process names are used in the output of various plotting functions. Both of these can also be set when the Gaussian process (list) is created by mlegp

Author(s)

Garrett M. Dancik [e-mail]

References

https://github.com/gdancik/mlegp/

See Also

mlegp for setting parameter and Gaussian process names during object creation

Examples

## fit multiple Gaussian processes to multiple observations ##
x = -5:5
y1 = sin(x) + rnorm(length(x),sd=.1)
y2 = sin(x) + 2*x + rnorm(length(x), sd = .1)
fitMulti = mlegp(x, cbind(y1,y2))

## plot diagnostics with default gp names ##
plot(fitMulti)

## change names and plot again ##
fitMulti = setGPNames(fitMulti, c("y1", "y2"))
plot(fitMulti)

## plot diagnostic for the first Gaussian process, predicted vs. parameter ##
plot(fitMulti[[1]], type = 2)

## change parameter names (of all Gaussian processes) and plot again ##
fitMulti = setParams(fitMulti, "param 1")
plot(fitMulti[[1]], type = 2)

mlegp documentation built on March 18, 2022, 5:29 p.m.