HybridForecaster: Forecast methods for covariate matrix
In yymmhaha/PackPaper1: Package related to my first paper

Description Usage Arguments Value Examples

Implements 3 types of forecasting methods for time series: single stands for auto.arima for each series individually; multi stands for VAR (vector autoregression) with default lag order 2; true stands for forecasting using the true X process specified.

1	HybridForecaster(X.fitting, horizon, datatype, parameters.of.X)

`X.fitting`	a matrix. The original training X matrix for forecast, DAT$X.for.fcst (10 series)
`horizon`	a number. Forecast horizon.
`datatype`	a string. Choose from 'simple', 'block1', 'block2'. Should be consistent with the true datatype used for simulation.
`parameters.of.X`	a list. The one that has been used for simulation.

a list of components

`single`	a matrix. Forecast values for each series produced by single method. Size should be horizon * nseries, 7 by 10
`multi`	a matrix. Forecast values for each series produced by multi method.
`true`	a matrix. Forecast values for each series produced by true method.