Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Nick Cox <njcoxstata@gmail.com> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | Re: st: Factor analysis and winsorizing/Outlier |
Date | Mon, 25 Nov 2013 12:37:45 +0000 |
See also previous discussions on this list e.g. http://www.stata.com/statalist/archive/2007-06/msg00241.html Nick njcoxstata@gmail.com On 25 November 2013 11:16, Maarten Buis <maartenlbuis@gmail.com> wrote: > The best thing to do is figure out why that observation is so special. > For example, your observations are hospitals, you mainly have regional > hospitals and your outlier is the only academic hospital. After that > you can make an informed decision on how best to include that > observation in your analysis. > > -- Maarten > > > On Mon, Nov 25, 2013 at 12:10 PM, Florian Christian Esser > <florian.esser@tu-dortmund.de> wrote: >> Hi Nick, >> thanks for your answer. >> What other things to are there? I am particularly worried about one >> outlier, that is in the range of "4" whereas most other results are based >> around "0". Sample size is 88. >> >> Thanks again >> Florian >> >>> I don't think there is any "should" here that comes on tablets of stone. >>> >>> If you are worried about outliers, there are many things you can do, >>> but if Winsorizing raw data is not enough to produce good results, >>> then further Winsorizing on what factor analysis produces is likely to >>> be going further down the wrong road. >>> Nick >>> njcoxstata@gmail.com >>> >>> >>> On 25 November 2013 08:26, Florian Christian Esser >>> <florian.esser@tu-dortmund.de> wrote: >>>> Hi everyone, >>>> >>>> I am using an exploratory factor analysis with three factors to derive a >>>> strategy construct based on secondary data. >>>> >>>> If I am winsorizing ma data, how should I proceed? Do I winsorize the >>>> input factors, do I winsorize the resulting construct, do I winsorize at >>>> both stages? >>>> >>>> Thanks a lot >>>> Florian >>>> >>>> * >>>> * For searches and help try: >>>> * http://www.stata.com/help.cgi?search >>>> * http://www.stata.com/support/faqs/resources/statalist-faq/ >>>> * http://www.ats.ucla.edu/stat/stata/ >>> * >>> * For searches and help try: >>> * http://www.stata.com/help.cgi?search >>> * http://www.stata.com/support/faqs/resources/statalist-faq/ >>> * http://www.ats.ucla.edu/stat/stata/ >>> >> >> >> * >> * For searches and help try: >> * http://www.stata.com/help.cgi?search >> * http://www.stata.com/support/faqs/resources/statalist-faq/ >> * http://www.ats.ucla.edu/stat/stata/ > > > > -- > --------------------------------- > Maarten L. Buis > WZB > Reichpietschufer 50 > 10785 Berlin > Germany > > http://www.maartenbuis.nl > --------------------------------- > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/faqs/resources/statalist-faq/ > * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/