VN ARMA

Description

Autoregressive model incorporating nonlinear regressions of component series.

Usage

1
2
VN.ARMA(variable, h = 1, Training_set = NULL, Seasonal_Factor = TRUE,
  Negative_Values = FALSE, Linear = TRUE, Dynamic = FALSE, s.t.n = 0.9)

Arguments

variable

Variable

h

Number of periods to forecast, defaults to 1.

Training_set

Sets the number of observations from 1:xx to monitor performance of forecast over in-sample range. Defaults to NULL to use all observations.

Seasonal_Factor

Automatically selects the best seasonal lag from the seasonality test. Defaults to TRUE. To use weighted average of all seasonal lags set to FALSE. Otherwise, input integer lag to use, i.e. Seasonal_Factor=12 for monthly data.

Negative_Values

If the variable can be negative, set to TRUE. Defaults to FALSE.

Linear

To use a linear regression of the component series, defaults to TRUE. To use a nonlineaer regression, set to FALSE.

Dynamic

To update the seasonal factor with each forecast point, set to TRUE. The default is FALSE to keep the original seasonal factor from the inputted variable for all forecasts.

s.t.n

Signal to noise parameter, sets the threshold of VN.dep which reduces "order" when order=NULL. Defaults to 0.9 to ensure high dependence for higher "order" and endpoint determination.

Value

Returns a vector of forecasts of length (h).

Author(s)

Fred Viole, OVVO Financial Systems

References

Viole, F. and Nawrocki, D. (2013) "Nonlinear Nonparametric Statistics: Using Partial Moments" http://amzn.com/1490523995

Examples

1
2
3
set.seed(123)
x<-rnorm(100)
VN.ARMA(x)