nl.acf: Autocorrelogram In weakARMA: Tools for the Analysis of Weak ARMA Models

 nl.acf R Documentation

Autocorrelogram

Description

Plots autocorrelogram for non linear process.

Usage

```nl.acf(
ar = NULL,
ma = NULL,
y,
main = NULL,
nlag = NULL,
conflevel = 0.05,
z = 1.2,
aff = "both"
)
```

Arguments

 `ar` Vector of AR coefficients. If `NULL`, we consider a MA process. `ma` Vector of MA coefficients. If `NULL`, we consider an AR process. `y` Univariate time series. `main` Character string representing the title for the plot. `nlag` Maximum lag at which to calculate the acf. If `NULL`, it is determinate by nlag = min(10log(n)) where n is the number of observation. `conflevel` Value of the confidence level, 5% by default. `z` Zoom on the graph. `aff` Specify the method between SN, M and both (see in Details).

Details

For the argument `aff` you have the choice between: `SN`, `M` and `both`. `SN` prints the self-normalized method (see Boubacar Maïnassara and Saussereau) in green, `M` prints the modified method introduced by Francq, Roy and Zakoïan (see also Boubacar Maïnassara) in red and `both` prints both of the methods.

Value

An autocorrelogram with every autocorrelations from 1 to a lag max, and with methods you choose to print.

Note

The only value available for the argument `conflevel` are 0.1, 0.05, 0.025, 0.01 or 0.005.

References

Boubacar Maïnassara, Y. 2011, Multivariate portmanteau test for structural VARMA models with uncorrelated but non-independent error terms Journal of Statistical Planning and Inference, vol. 141, no. 8, pp. 2961-2975.

Boubacar Maïnassara, Y.and Saussereau, B. 2018, Diagnostic checking in multivariate ARMA models with dependent errors using normalized residual autocorrelations , Journal of the American Statistical Association, vol. 113, no. 524, pp. 1813-1827.

Francq, C., Roy, R. and Zakoïan, J.M. 2005, Diagnostic Checking in ARMA Models with Uncorrelated Errors, Journal of the American Statistical Association, vol. 100, no. 470, pp. 532-544.

Lobato, I.N. 2001, Testing that a dependant process is uncorrelated. J. Amer. Statist. Assos. 96, vol. 455, pp. 1066-1076.

Examples

```est<-estimation(p = 1, q = 1, y = CAC40return.sq)
nl.acf(ar = est\$ar, ma = est\$ma, y = CAC40return.sq, main = "Autocorrelation of an ARMA(1,1)
residuals of the CAC40 return square", nlag = 20)

```

weakARMA documentation built on April 5, 2022, 1:16 a.m.