[R] Outlier Detection for timeseries

[Hans W. Borchers]

Pele <drdionc <at> yahoo.com> writes:

> 
> 
> Hello R users,
> 
> Can someone tell if there is a package in R that can do outlier detection
> that give outputs simiilar to what I got from SAS  below.
> 
> Many thanks in advance for any help!


I guess you are talking about the OUTLIER procedure in SAS that attempts
to detect 'additive outliers' and 'level shifts' in a 'response' series,
the second following Jong & Penzer's "Diagnosing shocks in time series".

I have not come across this method in R, but you might have a look into the
'robfilter' (Robust Time Series Filters) package with functions like
'dw.filter', 'adore.filter', or 'wrm.filter', see for instance

    "dw.filter is suitable for extracting low frequency components (the
    signal) from a time series which may be contaminated with outliers
    and can contain level shifts. For this, moving window techniques are 
    applied." 

If your time series is actually a response, you might prefer to look at
the series of residuals instead.


>                               Outlier Details
> 
> Approx
> Chi-    
> Prob>
>            Obs    Time ID         Type       Estimate      Square    
> ChiSq
> 
>             12       12.000000    Additive   2792544.6      186.13   
> <.0001
>             13       13.000000    Additive   954302.1       21.23   
> <.0001
>             15       15.000000    Shift      63539.3       
> 9.06    0.0026
>