Datastream US equity data quality

I recently came across an article that raised the issue of the quality of Datastream data. Ince & Porter (2006) warn researchers for potential flaws in the coverage by Datastream. The authors compared the US coverage of equity data to that of CRSP over the period 1975 and 2002. They found that certainly in the older years a large number of equities are missing in Datastream. An example is that for the year 1975 20% of common equity issues (found in the CRSP database) were missing from Datastream.

The problem is more important for small caps (= lower 20% of stocks measured by market capitalisation). Datastream also suffers to some extent from survivorship bias. The authors suggest a number of methods that alleviate the problems somewhat. More details are available in the article below:

Ince, O.S. & R.B. Porter, 2006, “Individual equity return data from Thomson Datastream: handle with care!,” Journal of Financial Research 29(4), 463-479.


%d bloggers like this: