| predict.kerfon {far} | R Documentation |
Computation of the prediction based on a functional kernel model
predict.kerfon(object, ..., newdata=NULL, label, na.rm=TRUE, positive=FALSE)
object |
A kerfon object result of the kerfon function. |
newdata |
A fdata object used in the
kerfon model to compute the prediction, or NULL
to predict one step forward with the data in object. |
label |
A vector of character giving the dates to associate to the predicted observations. |
na.rm |
Logical. Does the n.a. need to be removed. |
positive |
Logical. Does the result must be forced to positive values. |
... |
Additional arguments. |
This function computes one step forward prediction for a
kerfon model.
Use the newdata option to input the past values,
and the label option value to define the labels for the new
observations. Notices that the output as the same length as
newdata.
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 result will be maximum of 0 and the predicted value).
A fdata object.
J. Damon
# Simulation of a FARX process
data1 <- simul.farx(m=10,n=400,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)
# Cross validation
model1 <- kerfon(data=data1, x="X", r=10, na.rm=TRUE)
print(model1)
# Predicting values
pred1 <- predict(model1,newdata=select.fdata(data1,date=1:399))
# Persistence
persist1 <- pred.persist(select.fdata(data1,date=1:399),x="X")
# Real values
real1 <- select.fdata(data1,date=2:400)
errors0 <- persist1[[1]]-real1[[1]]
errors1 <- pred1[[1]]-real1[[1]]
# Norm of observations
summary(real1)
# Persistence
summary(as.fdata(errors0))
# kerfon model
summary(as.fdata(errors1))