rminer (version 1.4.6)

lforecast: Compute long term forecasts.

Description

Performs multi-step forecasts by iteratively using 1-ahead predictions as inputs

Usage

lforecast(M, data, start, horizon)

Arguments

M

fitted model, the object returned by fit.

data

training data, typically built using CasesSeries.

start

starting period (when out-of-samples start).

horizon

number of multi-step predictions.

Value

Returns a numeric vector with the multi-step predictions.

Details

Check the reference for details.

References

• This tutorial shows additional code examples: P. Cortez. A tutorial on using the rminer R package for data mining tasks. Teaching Report, Department of Information Systems, ALGORITMI Research Centre, Engineering School, University of Minho, Guimaraes, Portugal, July 2015. http://hdl.handle.net/1822/36210

• To check for more details: P. Cortez. Sensitivity Analysis for Time Lag Selection to Forecast Seasonal Time Series using Neural Networks and Support Vector Machines. In Proceedings of the IEEE International Joint Conference on Neural Networks (IJCNN 2010), pp. 3694-3701, Barcelona, Spain, July, 2010. IEEE Computer Society, ISBN: 978-1-4244-6917-8 (DVD edition). http://dx.doi.org/10.1109/IJCNN.2010.5596890

See Also

fit, CasesSeries, predict.fit, mgraph.

Examples

# NOT RUN {
ts=c(1,4,7,2,5,8,3,6,9,4,7,10,5,8,11,6,9)
d=CasesSeries(ts,c(1,2,3))
M=fit(y~.,d[1:7,],model="mlpe",search=2)
P1=predict(M,d[8:14,]) # single-step predictions
P2=lforecast(M,d,8,7) # multi-step predictions, horizon=7
print(mmetric(d$y[8:14],P1,"MAE"))
print(mmetric(d$y[8:14],P2,"MAE"))
L=vector("list",2); pred=vector("list",1);test=vector("list",1)
pred[[1]]=P1; test[[1]]=d$y[8:14]; L[[1]]=list(pred=pred,test=test,runs=1)
pred[[1]]=P2; test[[1]]=d$y[8:14]; L[[2]]=list(pred=pred,test=test,runs=1)
mgraph(L,graph="REG",Grid=10,leg=c("y","P1","P2"),col=c("black","cyan","blue"))
mgraph(L,graph="RSC",Grid=10,leg=c("P1","P2"),col=c("cyan","blue"))
# }