|
|
#1
16th November 2007, 09:55
|
|||
|
|||
CompoundingHappens Measurement Basics
This forum is for any questions or comments you have about the page
www.CompoundingHappens.com/meas_basics.htm. Questions, criticism, and counter-arguments are all welcome. Please try to state your case clearly and logically. The meas_basics.htm page on the CompoundingHappens.com web site presents information about some basic principles that apply to investment performance measurement. The forum is moderated by Damien Laker, the Founder and Principal of CompoundingHappens.com.
__________________
Damien Laker is one of the world's leading authorities in performance analytics. For more information visit Moderators Last edited by Damien Laker : 16th November 2007 at 09:59. Reason: fixed html links 
|
|
#2
22nd November 2007, 01:52
|
|||
|
|||
|
Question submitted by email
The following question was submitted by email:
I would like to calculate the equivalent of time weighted return for a short portfolio and a combined long/short portfolio. But I want to do so in a way that scores the performance of the basket of stocks independent of any cash position. The more typical situation which effectively holds the portfolio responsible for any unemployed cash is what I want to avoid. In the case of a long only book, it seems appropriate, for each day, to use the beginning of day market value plus any cash used to make purchases that day as the denominator, and to used the end of day market value plus the proceeds of any sales as the numerator in calculating the return for the day. In the cases of a short book and a combined long/short book, the equivalent approach is eluding me. Do you know how to go about this?
__________________
Damien Laker is one of the world's leading authorities in performance analytics. For more information visit Moderators
|
|
#3
22nd November 2007, 01:59
|
|||
|
|||
|
Answer to question submitted by email
To illustrate the answer to your question, let me assume that your equities are benchmarked against S&P500 index. A market neutral fund might have a benchmark of 100% cash, 100% S&P500 (for the longs), and -100% S&P 500 (for the shorts). This might sound silly, but the idea is that you measure the return of the long equities as one group, and the return of the short equities as another group. If the portfolio manager is doing a good job, the longs will outperform S&P 500, and the shorts will underperform S&P500 (remember: the benchmark weight for the shorts is negative).
I am assuming that you have read the stuff on my web site about calculating returns for short positions. The need to keep the longs and shorts separate was emphasised by Jose Menchero in his paper "Performance Attribution with Short Positions" (this may not be the exact title). If you separately measure the performance of the long positions, the short positions, and the cash positions (as three separate categories), you will be able to get a clear picture of how each of them is performing.
__________________
Damien Laker is one of the world's leading authorities in performance analytics. For more information visit Moderators
|
 |
| Thread Tools | Search this Thread |
| Display Modes | |
|
|
Linear Mode
