pred.persist: Forecasting using functional persistence

In far: Modelization for Functional AutoRegressive Processes

Forecasting using functional persistence

Description

Compute prediction of functional data using the persistence.

Usage

pred.persist(data, x, na.rm=TRUE, label, positive=FALSE)

Arguments

data	A fdata object.
x	A vector of character giving the names of the variables predicted.
na.rm	Logical. Does the n.a. need to be removed.
label	A vector of character giving the dates to associate to the predicted observations.
positive	Logical. Does the result must be forced to positive values.

Details

The persistence model is a beautiful way to name the simplest model ever. This model just suppose that the next observation will be equal to the previous one, that is to say, noting Xp(n) the prediction for X(n) that we "compute" :

Xp(n+1)=X(n)

Of course, the intrinsic purpose of this model is to be a comparison for more complicated models.

The x option is provided to select the variable to predict, using the label option value as the labels for the new observations. Notices that the output as the same length as the input as it is only a shift in time.

In some special context, the user may need to suppress the na.rm observations with the na.rm option, or force the prediction to be positive with the positive option (in this case the maximum of 0 and the past value is computed).

Value

A fdata object.

Note

This has been more instinctive to call this function predict.persist but, due to the naming mechanism introduced by the object oriented programming, this would have reefer to the predict method for the persist objects. As it isn't the meaning of this function, we preferred the name pred.persist.

Author(s)

J. Damon

See Also

predict.far,predict.kerfon.

Examples

  # Simulation of a FARX process
  data1 <- simul.farx(m=10,n=40,base=base.simul.far(20,5),
                base.exo=base.simul.far(20,5),
                d.a=matrix(c(0.5,0),nrow=1,ncol=2),
                alpha.conj=matrix(c(0.2,0),nrow=1,ncol=2),
                d.rho=diag(c(0.45,0.90,0.34,0.45)),
                alpha=diag(c(0.5,0.23,0.018)),
                d.rho.exo=diag(c(0.45,0.90,0.34,0.45)),
                cst1=0.0)
  print(data2 <- pred.persist(data1,x="X",label="41"))
  print(unclass(select.fdata(data1,date=paste(38:40)))$X)
  print(unclass(select.fdata(data2,date=paste(39:41))))

far documentation built on Aug. 14, 2022, 1:06 a.m.