forecast_function: forecast function that forecasts out how many steps you look...

In ryanbieber/Time-Series-Catch-All: Looks at time series data and produces the best forecast from them

Description

forecast function that forecasts out how many steps you look forward

Usage

forecast_function(
  orig_x,
  models,
  freq = freq,
  steps = steps,
  dlmPoly = dlmPoly,
  dlmSeas = dlmSeas,
  num.cores = num.cores,
  error = error,
  xreg = xreg,
  a.a.args = a.a.args,
  ets.args = ets.args,
  tbats.args = tbats.args,
  n.n.args = n.n.args
)

Arguments

orig_x	list of time series you want to forecast
models	list of models
freq	character string indicating "month" for monthly etc.
steps	integer that explains how far out you want forecast(needed for DLM)
dlmPoly	integer for the order of polynomial you want in dlm, default = 2
dlmSeas	integer for seasonal effect in dlm, default is 12 for monthly
num.cores	integer for the number of cores you want to use
error	either is "smape" or "mape", default is "mape"
xreg	list of matrix/array of exogenous regressors you want to use to forecast only applicable to certain models
a.a.args	list of arguments you want to change
ets.args	list of arguments you want to change
tbats.args	list of arguments you want to change
n.n.args	list of arguments you want to change

Value

list of models for models

Examples

## Not run: 
test <- replicate(5, list(ldeaths))
test_model <- par_time_series_catch(test, num.cores = 12)

## End(Not run)

ryanbieber/Time-Series-Catch-All documentation built on Oct. 13, 2020, 9:59 a.m.