Re: st: Detecting Outliers

[Ronnie Babigumira <rb.glists@gmail.com>]

I would go with -lv- or -iqr- and then list everything above the "severe" outlier fences and then crosscheck with the 
patient records. This is something I do in my own work. Indeed there will be some discussion on what to do if you find 
that they are not errors however, that was not your question.



Raphael Fraser wrote:

> my objective is simple to locate potential outliers then cross check
> with patient records to ensure that the information is legitimate.
> error free.
>
> On 5/2/06, Richard Goldstein <richgold@ix.netcom.com> wrote:
> > an outlier is something that needs to be predefined; I think
> > of an outlier as something that is surprising given my model
> > of the data; routines such as box plots force their model of
> > the data on you
> >
> > what is your model of the data?
> >
> > Rich
> >
> > Raphael Fraser wrote:
> > > I have 10 years data (5000 observations) on patients heights and
> > > weights. Is there any ado-file that could assist in locating possible
> > > outliers?
> > >
> > > Raphael
> >
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/support/faqs/res/findit.html
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/