[R] Stat question: How to deal w/ negative outliers?

S Ellison S.Ellison at LGCGroup.com
Fri Apr 12 19:39:35 CEST 2013


> -----Original Message-----
> I have a question:  I am using the interquantile method to 
> spot outliers & it gives me values of  say 234 & -120 or for 
> the higher & lower benchmarks.
> I don't have any issues w/ the higher end.  However I don't 
> have any negative values. My lowest possible value is 0. 
> Should I consider 0 as an outlier?
If your limits have been appropriately set and the lower outlier bound is -120, obviously not.

However, if your data are constrained positive and unusually small values are as likely to be erroneous as unusually large values, you probably should have set the bounds differently.

Try taking logs and see if that gives you a more or less symmetric distribution? 

S Ellison

This email and any attachments are confidential. Any use...{{dropped:8}}

More information about the R-help mailing list