Re: removing autocorrelation

Thanks. In the case of monthly data (annual cycle removed) and
annual data,
there is no real trend in the data but it has a significant lag 1
autocorrelation.

Mike

On Jan 17, 2007, at 3:05 PM, Dennis Shea wrote:

>> Does anyone know a technique in NCL to remove the temporal
>> autocorrelation of a time series?
>> Thanks, Michael
>> _______________________________________________
>
>
> There is no built in autoregressive model in NCL. Usually,
> the following is sufficient.
>
> [1] detrend the data [remove linear contribution to autocorrelation]
>
> [2] if monthly compute the 12 monthly climatologies
> [remove annual cycle]
>
> http://www.ncl.ucar.edu/Document/Functions/Contributed/
> clmMonTLL.shtml
> clmMonTLLL
>
> then calculate the anomalies
>
> http://www.ncl.ucar.edu/Document/Functions/Contributed/
> calcMonAnomTLL.shtml
>
> calcMonAnomTLLL
>
> [3] calculate r1 ... Is it significant? If not, the data
> are considered indpendent. If it is still significant,
> more sophisticated methods may be required.
> ---
>
> daily
>
> http://www.ncl.ucar.edu/Document/Functions/Contributed/clmDayTLL.shtml
> http://www.ncl.ucar.edu/Document/Functions/Contributed/
> calcDayAnomTLL.shtml
>
> good luck
> D
>
> _______________________________________________
> ncl-talk mailing list
> ncl-talk_at_ucar.edu
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk

_______________________________________________
ncl-talk mailing list
ncl-talk_at_ucar.edu
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Wed Jan 17 2007 - 14:16:20 MST