Many papers, e.g. in The Journal of Finance, discuss DGTW adjusted returns (or DGTW abnormal returns) instead of just returns.
What are these and how does one compute them?
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
5
Following Daniel, Grinblatt, Titman, and Wermers (1997) "D.G.T.W.!", DGTW subtracts from each stock return the return on a portfolio of firms matched on market equity, market-book, and prior one-year return quintiles.
Daniel, K., Grinblatt, M., Titman, S., Wermers, R., 1997. Measuring mutual fund performance with characteristic-based benchmarks, Journal of Finance 52, 1035–1058.
The DGTW paper tries to decide whether stock funds are good in picking stocks and timing the market. My understanding is that the intended application of DGTW returns is to have a criterion for stocks which outperfrom its benchmark and then make conclusions about the stock-picking abilities of the fund manager. Of course the same principle applies for other stock picking entities, such as computer algorithms.
Edit: you can download the benchmark returns from the homepage of one of the authors.
-
$\begingroup$ Thanks. What is the argument for making such a correction? First thing that comes to mind is to be able to compare similar assets. $\endgroup$– SunvJan 24, 2014 at 15:10
-
$\begingroup$ @Anna you might want to incorporate that comment into the answer. $\endgroup$– JohnJan 24, 2014 at 17:32
-
$\begingroup$ @John thx for the tip. done. $\endgroup$– user1157Jan 24, 2014 at 17:36
-
$\begingroup$ @Anna : in some applications economists use DGTW returns without the purpose of looking at the performance of fund managers, e.g. this article p. 22: www3.nd.edu/~pgao/papers/Google_JF2011.pdf. Why do you think they use DGTW returns? $\endgroup$– SunvMar 6, 2014 at 11:07
-
$\begingroup$ They are testing a stock picking criterion related to search term frequencies, you can consider this as a "non-human stock picker". Still the same principle applies. $\endgroup$– user1157Mar 6, 2014 at 11:11