Selection of Subset ARMA Models
This function finds a number of subset ARMA models. A "long" AR model is fitted to the data y to compute the residuals which are taken as a proxy of the error process. Then, an ARMA model is approximated by a regression model with the the covariates being the lags of the time series and the lags of the error process. Subset ARMA models may then be selected using the subset regression technique by leaps and bounds, via the regsubsets function of the leaps package in R.
armasubsets(y, nar, nma, y.name = "Y", ar.method = "ols", ...)
y |
time-series data |
nar |
maximum AR order |
nma |
maximum MA order |
y.name |
label of the time series |
ar.method |
method used for fitting the long AR model; default is ols with the AR order determined by AIC |
... |
arguments passed to the plot.armasubsets function |
An object of the armasubsets class to be processed by the plot.armasubsets function.
Kung-Sik Chan
set.seed(92397) test=arima.sim(model=list(ar=c(rep(0,11),.8),ma=c(rep(0,11),0.7)),n=120) res=armasubsets(y=test,nar=14,nma=14,y.name='test',ar.method='ols') plot(res)
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.