[R] Time series (trend over time) for irregular sampling dates and multiple sites

[Abs Spurdle]

> > My data has a few problems: (1) I think I will need to fix the effects
of
> > seasonal variation (Monthly) and (2) of possible spatial correlation
> > (probability of finding an item is higher after finding one since they
can
> > come from the same ship). (3) How do I handle the fact that the
> > measurements were not taken at a regular interval?
>
> Can I ask two questions:
> (1) Is the data autocorrelated (or "Seasonal") over time?
> If not then this problem is a lot simpler.
> (2) Can you expand on the following statement?
> "possible spatial correlation (probability of finding an item is higher
after finding one since they can come from the same ship"

I just had a closer look at your example.

You've tried to model nMonth (presumably in {1, 2, ..., 12}) but is there a
long term trend, over Year?
Also, I'm not an expert on mgcv, but I was wondering if you want bs="cp"
rather than bs="ps"?

When you say "measurements were not taken at a regular interval" are you
referring to the variable "DaysIa"?
In which case, my previous question about autocorrelation applies to this
variable.

	[[alternative HTML version deleted]]