[R-sig-ME] Temporal autocorrelation when no significant time trend

[Ben Bolker]

Karel Viaene <Karel.Viaene at ...> writes:

> I have a short question: Can I still add a temporal autocorrelation  
> structure to a GAMM when the time variable is not significant, so when  
> there is no significant time trend?

  I don't see why not.  Did you try it?  Were the answers sensible?
Did you get answers reasonably similar to those you would get by
just looking at the ACF of the model residuals?