Wednesday, May 12, 2010

Is the math REALLY that difficult?

I'm conducting a GIPS(R) (Global Investment Performance Standards) verification and encountered something somewhat unusual. The client uses a well known software package to calculate their returns and use a daily method. Here are the approximate details for one day that is behind this post:
  • Beginning market value = $327,000
  • Cash flow = -$337,000
  • Ending market value = $940.
Obviously, the portfolio grew in value during the day which allowed them to distribute more money than they started the day with. Okay, so what return does the system yield? Actually it doesn't provide ANY return and simply indicates in a footnote that a return isn't possible. BUT, why not? We can employ a simple Modified Dietz formula:

which will yield a return of 3.35%. Is it really that hard? I don't think so. Perhaps you would think so at first glance, but it really isn't.


  1. So, by your formula (which assumes the cash flow is at the beginning of the day):
    940 - 327,000 - (-337,000) = 10,940 and
    327,000 +(-337,000) = -10,000
    10,940/-10,000 = -1.094

    But your answer is actually (EMV-BMV-CF)/BMV which assumes the cash flow is at the end of the day.

    I guess it REALLY is that hard.

  2. Thanks for checking my math. No, it's not hard ... only when we make Excel mistakes. I should have realized there was an error because we generally recommend that outflows be treated as end-of-day events. I've adjusted the formula to show the proper form. When there are very large flows, the problem with the start-of-day approach clearly surfaces, as it did here.

  3. Hurrah! The vendor recognized that they should treat inflows as start-of-day and outflows as end-of-day events. They will remain unnamed, but it's great to hear that yet another vendor has seen the benefit of such a switch.

  4. This post also make me realized we have a problem in our calculation for so long..thanks !


Note: Only a member of this blog may post a comment.