GPFDA (version 1.1)

gppredict: Prediction of the Gaussian Process

Description

Predict the new points in Gaussian Process using the training results or manual input

Usage

gppredict(train=NULL,Data.new=NULL,hyper=NULL, Data=NULL, Y=NULL, Cov=NULL,
  gamma=NULL,lrm=NULL,mean=0)

Arguments

train

The result from training. Default to be NULL, if not NULL, other arguments (except for Data.new) will replaced by NULL.

Data.new

The test data. Must be a vector or a matrix.

hyper

Hyper-parameter estimated from training. Can use manual input. Default to be NULL.

Data

The data from training. Must be a vector or a matrix. Default to be NULL.

Y

The response from training. Must be a vector or a matrix. Default to be NULL.

Cov

Names of covariance functions used. Default to be NULL.

gamma

Parameter used in power exponential covariance function. Default to be NULL.

lrm

The linear trend from learning. Default to be lrm. If lrm exists from learning, NULL will be replaced.

mean

Is the mean taken out when analysis? Default to be 0, which assumes the mean is zero. if assume mean is a constant, mean=1; if assume mean is a linear trend, mean='t'.

Value

• CovFun
  Covariance function type
• fitted
  Fitted value of training data
• fitted.sd
  Standard deviation of the fitted value of training data
• gamma
  Parameter used in powered exponential covariance function
• hyper
  Hyper-parameter estimated from training data
• I
  Variance of the estimated hyper-parameters
• mu
  Estimated prediction mean
• sigma
  Estimated prediction variance
• train.x
  Training covariates
• train.y
  Training response

Details

Use the result from training to predict the value for new points.

References

Shi, J Q., and Choi, T. (2011), Gaussian Process Regression Analysis for Functional Data, Springer, New York.

See Also

gpr

Examples

library(GPFDA)
library(MASS) ## used to generate data
hp <- list('pow.ex.w'=log(10),'linear.a'=log(10),'pow.ex.v'=log(5),'vv'=log(1))
c <- seq(0,1,len=40)
idx <- sort(sample(1:40,21))
X <- as.matrix(c[idx])
Y <- (mvrnorm(n=40,mu=c-c,Sigma=(cov.linear(hp,c)+cov.pow.ex(hp,c)))[,1])+sin(c*6)
Y <- as.matrix(Y[idx])
x <- as.matrix(seq(0,1,by=0.03))
a <- gpr(X,Y,c('linear','pow.ex'))
b <- gppredict(a,x)