predtoolsTS (version 0.1.1)

modl.dataMining: Train the data

Description

Train the time serie(as data frame) to build the model.

Usage

modl.dataMining(form, tserieDF, algorithm, timeControl, metric = "RMSE",
  maximize = FALSE)

Arguments

form

A formula of the form y ~ x1 + x2 + ...

tserieDF

Data frame.

algorithm

A string. Algorithm to perform the training. Full list at http://topepo.github.io/caret/train-models-by-tag.html. Only regression types allowed.

timeControl

trainControl object.

metric

A string. Specifies what summary metric will be used to select the optimal model. Possible values in caret are "RMSE" and "Rsquared". "RMSE" set as default. If you used a custom summaryFunction(see ?trainControl) your metrics will prevail over default.

maximize

A logical. Should the metric be maximized or minimized? Default is FALSE, since that is what makes sense for time series.

Value

train object

Examples

# NOT RUN {
modl.dataMining(form=Class ~ .,
 tserieDF=modl.tsToDataFrame(AirPassengers,formula=c(1:20)),
 algorithm='rpart',
 timeControl=modl.trControl(initialWindow=30,horizon=15,fixedWindow=TRUE))
# }