R help - Nov 2009 - Why arima does not work when the time series is a costant?

Hi! 

One of my time series happens to be a constant. When I call arima with c(0,
0, 0), it gives me error. Here is an example:

ts1 <- ts(rep(1, 29))
fit <- arima(ts1, order=c(0,0,0))

When I run the second line above, it gives me the following error
"Error in decompose(ts(x[1L:wind], start = start(x), frequency = f),
seasonal) : 
  time series has no or less than 2 periods" 

I was confused why arima does not work if my time series is a constant. It
should be very simple - resulted model should be a constant with the value
of 1. 

Thanks,
Aijun.

-- 
View this message in context:
http://old.nabble.com/Why-arima-does-not-work-when-the-time-series-is-a-costant--tp26287374p26287374.html
Sent from the R help mailing list archive at Nabble.com.

	[[alternative HTML version deleted]]