ttsCaret: Train time series by 'caret' and produce two types of time...

In iForecast: Machine Learning Time Series Forecasting

Train time series by caret and produce two types of time series forecasts: static and recursive

Description

It generates both the static and recursive time series plots of machine learning prediction object generated by package caret.

Usage

ttsCaret(
  y,
  x=NULL,
  method,
  train.end,
  arOrder=2,
  xregOrder=0,
  type,
  tuneLength =10,
  preProcess = NULL,
  resampling="boot",
  Number=NULL,
  Repeat=NULL)

Arguments

y	The time series object of the target variable, or the dependent variable, with timeSeries or zoo format, must have dimension. y can be either binary or continuous.Time format must be "
x	The time series matrix of input variables, or the independent variables, with timeSeries or zoo format.Time format must be "
method	The train_model_list of caret. While using this, make sure that the method allows regression. Methods in c("svm","rf","rpart","gamboost","BstLm","bstSm","blackboost") are feasible.
train.end	The end date of training data, must be specificed.The default dates of train.start and test.end are the start and the end of input data; and the test.start is the 1-period next of train.end.
arOrder	The autoregressive order of the target variable, which may be sequentially specifed like arOrder=1:5; or discontinuous lags like arOrder=c(1,3,5); zero is not allowed.
xregOrder	The distributed lag structure of the input variables, which may be sequentially specifed like xregOrder=0:5; or discontinuous lags like xregOrder=c(0,3,5); zero is allowed since contemporaneous correlation is allowed.
type	The additional input variables. We have four selection: "none"=no other variables, "trend"=inclusion of time dummy, "season"=inclusion of seasonal dummies, "both"=inclusion of both trend and season. No default.
tuneLength	The same as the length specified in train function of package caret.
preProcess	Whether to pre-process the data, current possibilities are "BoxCox", "YeoJohnson", "expoTrans", "center", "scale", "range", "knnImpute", "bagImpute", "medianImpute", "pca", "ica" and "spatialSign".The default is no pre-processing.
resampling	The method for resampling, as trainControl function list in package caret. The default is "boot" for bootstrapping with 25 replications. Current choices are c("cv","boot","repeatedcv","LOOCV") where "cv" is K-fold CV with a default K=10 or specified by the "Number" below, "LOOCV" denotes the leave-one-out CV
Number	The number of K for K-Fold CV, default (NULL) is 10; for "boot" option, the default number of replications is 25
Repeat	The number for the repeatition for "repeatedcv".

Details

This function calls the train function of package caret to execute estimation. When execution finished, we compute two types of time series forecasts: static and recursive.

Value

output	Output object generated by train function of caret.
arOrder	The autoregressive order of the target variable used.
data	The dataset of imputed.
dataused	The data used by arOrder, xregOrder, and type.
training.Pred	All tuned prediction values of training data, using besTunes to extract the best prediction.

Author(s)

Ho Tsung-wu <[e-mail]>, College of Management, National Taiwan Normal University.

Examples

# Cross-validation takes time, example below is commented.
## Machine Learning by library(caret)
library(zoo)
#Case 1. Low frequency
data("macrodata")
dep <- macrodata[569:669,"unrate",drop=FALSE]
ind <- macrodata[569:669,-1,drop=FALSE]
train.end <- "2018-12-01"# Choosing the end dating of train

models <- c("glm","knn","nnet","rpart","rf","svm","enet","gbm","lasso","bridge")[2]
type <- c("none","trend","season","both")[1]
Caret <- ttsCaret(y=dep, x=NULL, arOrder=c(1), xregOrder=c(1),
 method=models, tuneLength =1, train.end, type=type,
 resampling=c("boot","cv","repeatedcv")[2],preProcess = "center")
 testData1 <- window(Caret$data,start="2019-01-01",end=end(Caret$data))
P1 <- iForecast(Model=Caret,newdata=testData1,type="static")
P2 <- iForecast(Model=Caret,newdata=testData1,type="dynamic")

tail(cbind(testData1[,1],P1,P2))

#Case 2. High frequency
#head(ES_15m)
#head(ES_Daily)
#dep <- ES_15m #SP500 15-minute realized absolute variance
#ind <- NULL
#train.end <- as.character(rownames(dep))[as.integer(nrow(dep)*0.9)]

#models<-c("svm","rf","rpart","gamboost","BstLm","bstSm","blackboost")[1]
#type<-c("none","trend","season","both")[1]
# Caret <- ttsCaret(y=dep, x=ind, arOrder=c(3,5), xregOrder=c(0,2,4),
# method=models, tuneLength =10, train.end, type=type,
# resampling=c("boot","cv","repeatedcv")[2],preProcess = "center")
#testData1<-window(Caret$data,start="2009-01-01",end=end(Caret$data))
#P1<-iForecast(Model=Caret,newdata=testData1,type="static")
#P2<-iForecast(Model=Caret,newdata=testData1,type="dynamic")

iForecast documentation built on July 10, 2023, 1:59 a.m.