global portfolio return attribution

You may also like...

9 Responses

  1. Stephanie says:

    Okay, I *think* I managed to follow that… there’s a reason I studied biology in college! 😛 Oh, and I loved your little comment at the end. Way to go, kiddo! 🙂

  2. Max says:

    Thanks so much for posting this. After going through it a few times I’ve finally gotten this as well as the micro performance attribution formula and implementation formula down for good.

    I’m a very visual learner so this helped a lot. Do you have any other tricks like this for the curriculum?

    • John says:

      No more slideshows I’m afraid, although if I should have the displeasure of having to study this curriculum next year, I’ll probably come up with a few more to help myself learn the material. Glad it helped.

  3. Jaideep says:

    DELIGHTFUL easter egg on last slide…a picture of your kiddo with a note that says “my 19 mo. old learned how to count to six while I built this spreadsheet” puts the whole study push into perspective. Thanks for sharing!!!!

    Your diagram helps immensely. I think there is room to clarify further with terminology here…

    “DECOMPOSITION” needs no benchmark, it takes a portfolio return and chops it into “COMPONENTS” . A clean example of this is formula (4) on page 209.

    “ATTRIBUTION” on the other hand needs benchmark and chops up the differential return (the value added return) into “CONTRIBUTIONS” A clean example of this is formula (16) on page 162.

    The big formula (5) on page 211 combines COMPONENTS and CONTRIBUTIONS (i.e. DECOMPOSITION and ATTRIBUTION)…not sure why they do this mixed treatment. This to me was very confusing.

    Also the CFAI examples ask to DECOMPOSE into Cap. gain, yield, and currency “COMPONENTS” unfortunately the EOC pollute the vocabulary.

    I think there is value in keeping the vocabulary clear, to know exactly what is being discussed.

    • John says:

      Thanks for the encouragement, and you’re very right regarding attribution vs. decomposition. I understood that distinction better after I built the slideshows. The thing to keep clear is that in an attribution, all the terms besides the benchmark return can still be thought of as a decomposition. I doubt the terminology will be that big of an issue on the exam, but who knows.

  4. Dilip says:

    In case of Calculating the Performance Attribution how can we do it for multi period if the weight is known on daily basis.

    • John says:

      That’s a good subject for discussion over at, but my first reaction is that it doesn’t matter what “periods” the attribution is using in terms of the formulas. In your case you would be attributing daily performance. I shudder a little bit as a portfolio manager to think what kinds of decisions are made on the basis of daily performance attribution. You could take a weighted averages and do a calculation over a longer period.

  5. C P Krishnan says:

    Hi John

    Had a small favor to ask. I know you had published the “Sector-Allocation” Attribution model graph someplace – but later it got overwritten by the “Global attribution” one.

    Would you know if you had a copy someplace and could send the same to me?

    C.P.Krishnan (cpk123 on AF).

  6. John says:

    Honestly, I’m not sure what you’re talking about; I’m pretty sure these two were the only Powerpoint presentations I’ve ever built. Maybe you can refresh my memory on what the formula is, or else point me to a thread on AF where I refer to the one you’re talking about. I know you can take the second one (Multi-period performance attribution) and read “market allocation” as “sector allocation” instead of “country allocation”, but I doubt that’s what you’re referring to.

Leave a Reply

%d bloggers like this: