R: Generate a tsibble of synthetic data from a Mixture...

generate.mar

R Documentation

Generate a tsibble of synthetic data from a Mixture Autoregressive model

Description

This function simulates multiple random sample paths from a mixture of k Gaussian AR(p) processes. The model is of the form

y_t = φ_{0,i} + φ_{1,i}y_{t-1} + … + φ_{p,i}y_{t-p} + σ_{i,t}ε_t

with probability α_i, where ε_t is a N(0,1) variate. The index of the tsibble is guessed from the MAR model seasonal periods.

Usage

## S3 method for class 'mar'
generate(x, length = 100, nseries = 10, ...)

## S3 method for class 'ets'
generate(x, length = 100, nseries = 10, ...)

## S3 method for class 'Arima'
generate(x, length = 100, nseries = 10, ...)

Arguments

`x`	A 'mar' object, usually the output of `mar_model()`.
`length`	length of series to generate
`nseries`	number of series to generate
`...`	Other arguments, passed to `simulate.mar`.

Value

'tsibble' object with 'length' rows and 3 columns.

Author(s)

Rob J Hyndman

References

Feng Li, Mattias Villani, and Robert Kohn. (2010). Flexible Modeling of Conditional Distributions using Smooth Mixtures of Asymmetric Student T Densities, Journal of Statistical Planning and Inference, 140(12), pp. 3638-3654.

Examples

# MAR model with constant variances
phi <- cbind(c(0, 0.8, 0), c(0, 0.6, 0.3))
weights <- c(0.8, 0.2)
model1 <- mar_model(phi = phi, sigmas = c(1, 2), weights = weights)
generate(model1, nseries = 5)
# MAR model for hourly data with daily and weekly periods
hourly_model <- mar_model(seasonal_periods = c(24, 24*7))
generate(hourly_model)