| LB.test {TSA} | R Documentation |
Portmanteau Tests for Fitted ARIMA models
Description
This function modifies the Box.test function in the stats package, and it computes the Ljung-Box or Box-Pierce tests checking whether or not the residuals appear to be white noise.
Usage
LB.test(model, lag = 12, type = c("Ljung-Box", "Box-Pierce"), no.error = FALSE,
omit.initial = TRUE)
Arguments
model |
model fit from the arima function |
lag |
number of lags of the autocorrelation of the residuals to be included in the test statistic. (default=12) |
type |
either Ljung-Box or Box-Pierce |
no.error |
a system variable; normally it is not changed |
omit.initial |
if true, (d+Ds) initial residuals are omitted from the test |
Value
a list:
statistics |
test statistic |
p.value |
p-value |
parameter |
d.f. of the Chi-square test |
lag |
no of lags |
Author(s)
Kung-Sik Chan, based on A. Trapletti's work on the Box.test function in the stats package
References
Box, G. E. P. and Pierce, D. A. (1970), Distribution of residual correlations in autoregressive-integrated moving average time series models. Journal of the American Statistical Association, 65, 15091526.
Ljung, G. M. and Box, G. E. P. (1978), On a measure of lack of fit in time series models. Biometrika 65, 553564.
Examples
data(color)
m1.color=arima(color,order=c(1,0,0))
LB.test(m1.color)