How subjective is data cleaning in finance research?
Posted: 05 June 2015 02:01 PM   [ Ignore ]
Newbie
Rank
Total Posts:  5
Joined  2015-06-05

Hello all,

I’ve always have this question when I am reading papers in corporate finance. some filters are the same in almost all papers (i.e. they all eliminate financials and utilities or set R&D equal to zero when missing). however, sometimes I see that some papers delete some observations if a variables (i.e. assets-total) is missing. or if it is negative or zero.

Is there a guideline about this practice? or in general about data cleaning process in financial research?

Regards,

Profile
 
 
Posted: 05 June 2015 02:11 PM   [ Ignore ]   [ # 1 ]
Administrator
Avatar
RankRankRankRank
Total Posts:  901
Joined  2011-09-19

hi Moh,

There are some general characteristics like you point out (regulated industries), or firm size (very small firms have low profitability and (relatively) high valuations, so easily introduce noise when mixed with ‘normal’ firms.

It would also depend on the ‘seminal’ work in the research area. If the initial work does filtering that doesn’t make sense, follow up papers will ‘need to’ do that as well, just because. Same goes for choice of control variables, by the way.

If there are written guidelines I would be interested in getting a copy smile

Best,

Joost

 Signature 

To reply/post new questions: Please use the group WRDS/SAS on Google Groups! http://groups.google.com/d/forum/wrdssas

Profile
 
 
Posted: 05 June 2015 04:37 PM   [ Ignore ]   [ # 2 ]
Newbie
Rank
Total Posts:  5
Joined  2015-06-05

Joost,

Then for each project, I will focus on the most seminal paper in that area step by step and cite them for that methodology.

Thanks again!

Best,
Moh

Profile
 
 
Posted: 05 June 2015 07:21 PM   [ Ignore ]   [ # 3 ]
Administrator
Avatar
RankRankRankRank
Total Posts:  901
Joined  2011-09-19

Sounds like a plan smile

Best,

Joost

 Signature 

To reply/post new questions: Please use the group WRDS/SAS on Google Groups! http://groups.google.com/d/forum/wrdssas

Profile