I keep reading posts about Compustat database being superior compared to Reuters.
Why are we so sure about that? Does Computstat really add value for us quantative investors by its ‘smart’ interpretation of filings?
This paper published in 2005 comes to a different conclusion and suggests that Reuters Worldscope should give better results for quantative investing.
[i]The overall conclusion is that:
For U.S. firms Worldscope is as competitive as Compustat even though nobody uses Worldscope for research with U.S. data.
We find no statistical or methodological reason why Worldscope should not be used for research.
The number of firms and the coverage of variables is similar.
Applied to multiple valuation the results are different. One can see a significantly lower valuation error for Worldscope independently of the number of firms in the underlying dataset.[/i]
Reuters for current data is good, and possibly better due to slightly deeper coverage and more line items (although hard to use)
But the database ability to recreate the past is not even close. Only reason we had semi decent Point in Time data with Reuters is because we downloaded it every day for 10 years. Everything Compustat does allows the past to be re-created. Reuters loses some connections between dead shares and parent company, and overwrites a ton of stuff )like preliminary data and industry taxonomy).
Since we can’t have the best of both worlds, Compustat is best for us. Reuters would be number two from all the rest (mergent, morningstar, factset, zacks)
I understand the case for using a Point in Time solution vs a non Point in Time solution as managing 10+ years of weekly snapshot data seems very cumbersome.
But what exactly is the case for Compustat Point in Time vs Reuters Point in Time?
Just from my personal experiences, I noticed that while using the Reuters data feed sometimes my simulation holdings as of a specific date changes when run at different time with same settings. That could mean, for some reason, the historical data was changed. However, with the new data feed, I see fewer such events.
