var_ar: Variance of an AR process

In JuGross/ensAR: Autoregressive postprocessing methods for ensemble forecasts

Description

Comptutes the variance of an autoregressive (AR) process.

Usage

var_ar(ar = numeric(), i_var = 1)

Arguments

ar	A vector of autoregression coefficients.
i_var	The innovations variance.

Details

The variance of an AR process of order p is

γ_{0} = \frac{σ^2}{1 - φ_{1} ρ_{1} - φ_{2} ρ_{2} - \cdots - φ_{p} ρ_{p}}

where σ^2 is the variance of the innovations and φ_{j} are the autoregression coefficients. The autocorrelations ρ_{j} are computed by ARMAacf.

Value

The variance of the process.

Author(s)

J. Gross, A. Moeller.

Examples

## x <- arima.sim(list(ar = 0.7), sd = 0.5, 100000)
## var(x)
var_ar(ar = 0.7, i_var= 0.5^2)

JuGross/ensAR documentation built on May 10, 2019, 8:23 a.m.