robust_starting_point: roubst_starting_point

In Mufabo/Rrobustsp: Robust Signal Processing

roubst_starting_point

Description

The function robust_starting_point(x,p,q) provides a robust initial estimate for robust ARMA parameter estimation based on BIP-AR(p_long) approximation. It also computes an outlier cleaned signal using BIP-AR(p_long) predictions

Usage

robust_starting_point(x, p, q, recursion_num = 0)

Arguments

x:	numeriv vector, the time series / signal
p:	AR order
q:	MA order
recursion_num:	integer. If parameter estimation is instationary, run robust_starting_point recursively on the filtered signal.

Value

beta_inital: ARMA coefficient estimate and the intercept

cleaned_signal: cleaned signal

Examples

data(x_ao_arma)
library(pracma)

robust_starting_point(x_ao_arma, 1, 1)
#$beta_initial
#ar1         ma1   intercept
#0.81314423  0.14617580 -0.05612576

Mufabo/Rrobustsp documentation built on June 11, 2022, 10:41 p.m.