GPFDA (version 1.1)

gpfrpred: Prediction of the Gaussian Process using functional regression

Description

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

Usage

gpfrpred(object,data=NULL,newtime=NULL,data.new = NULL,type=1,yregfd=NULL, 
  hyper.p=NULL, iuu=NULL, Cov=NULL,gamma=1)

Arguments

object

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

data

The test data. Must have the format from wrap. For type 1 prediction. If `newtime' is missing, but type is 2, time in data will be used for type 2 prediction.

newtime

The new time point to predict. For type 2 prediction.

data.new

The test data for predict. For type 2 prediction.

type

Prediction types. See details below.

yregfd

Functional coefficients.

hyper.p

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

iuu

Scaler variables for prediction. Always required for both type of predictions.

Cov

Names of covariance functions used. Default to be NULL.

gamma

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

Value

• ypred
  matrix of predicted value with confidence interval
• time
  new time points
• y_gppred
  predicted value
• object
  all items trained from gpfr if exists

Details

Two types of prediction are supplied. Type one is the new batch has a few observations, type two is the new batch has no observations.

References

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

See Also

gpr