ts_generate: Generate time series with various length by GRATIS.

In xqnwang/fuma: Forecast uncertainty based on model averaging (FUMA)

Description

Generate time series with certain seasonal period and length according to GRATIS proposed by Kang et al. (2019).

Usage

ts_generate(
  n.ts = 1,
  freq = 1,
  nComp = NULL,
  length = NULL,
  h = 1,
  parallel = FALSE,
  num.cores = 2
)

Arguments

n.ts	number of time series to be generated.
freq	seasonal period of time series to be generated. For example, freq of the yearly, quarterly, and monthly series are 1, 4, and 12, respectively.
nComp	number of mixing components when simulating time series using MAR models.
length	length vector of time series to be generated. The length of the generated time series is simply sampled from the length vector.
h	forecasting horizon of the generated time series.
parallel	logical. If TRUE then the generation process is conducted in parallel.
num.cores	the specified amount of parallel processes to be used if parallel = TRUE.

Value

A list of the generated time series.

References

Kang et al. (2019).

Examples

 x_y <- ts_generate(n.ts = 10, freq = 4, nComp = 2, length = c(20,25,30), h = 8)
 x_y$Q1$pars
 forecast::autoplot(x_y$Q1$x)

xqnwang/fuma documentation built on May 29, 2021, 6:38 a.m.