I am currently designing an attribution model for a client that invests in options, alongside their equities. I must confess that putting this together with their data is taking a bit longer than I had planned, but I am hopeful we will present them with a valuable tool they can use with their clients and prospects.
The general approach is to segregate the options from the equities. And since these are covered writes, the covering securities or cash will be with the options.
The result will be a clean portrayal of where the return comes from.
More details will follow, most likely in The Spaulding Group's monthly newsletter. So stay tuned!
.png)
David,
ReplyDeleteI agree that derivatives may be treated as a separate asset class in an appropriately versatile decision attribution model. I also agree that one may treat the ‘covering securities or cash’ as belonging to that ‘derivative’ asset class. (I do not believe that this has anything special to do with the extent of overlap with the benchmark.)
And I also agree with you that there are critical subtleties involved regarding this segregation, especially in the presence of trades between elements of one asset class and another asset class.
At Opturo, we worked very hard to get all these aspects of decision trees implemented in an appropriate manner. We found it takes being very clear and consistent regarding the questions about ‘where the return comes from’ that each type of Attribution model intends to address and then to address them precisely.
Andre
Andre, thanks for your comments. Please consider writing an article that describes your approach for The Journal of Performance Measurement. The industry can always benefit from learning new approaches.
ReplyDelete