all_in_one_time_series: God version of prediction

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

Description

God version of prediction

Usage

all_in_one_time_series(
  original,
  freq = "month",
  steps = 3,
  dlmPoly = 2,
  dlmSeas = 12,
  num.cores = 2,
  error = "mape",
  xreg = NULL,
  a.a.args = list(NULL),
  ets.args = list(NULL),
  tbats.args = list(NULL),
  n.n.args = list(NULL),
  m = 10
)

Arguments

original	list of time series you want to forecast
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	error method you want to use for backtesting
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
m	how many times you would like it to impute values

Value

a data frame with your series original values plus the forecasts with added steps

Examples

## Not run: 
all_in_one_time_series(mtcars)

## End(Not run)

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