-----Original Message-----
From: guillaume.nicoulaud at halbis.com
[mailto:guillaume.nicoulaud at halbis.com]
Sent: Wednesday, 10 January, 2007 15:04
To: Christian Prinoth
Cc: r-sig-finance at stat.math.ethz.ch;
r-sig-finance-bounces at stat.math.ethz.ch
Subject: R?f. : Re: [R-SIG-Finance] solve.QP (for portfolio
optimization)
--- about Christian's two-step approach ---
I've tryed this. The issue is that provided we want to
minimize the *overall* portfolio variance we have to manage
potential correlations between longs and shorts (e.g. if some
of your longs are "anti-correlated" - is that the correct word?
- with some of your shorts). In my tests, simple equally
weighted portfolios are less volatiles.
--- Brian wrote ---
" Markowitz style optimization will try to minimize variance
across the entire portfolio. You *want* the short portfolio
to decline in value, as much as possible. "
I agree. Yet I could think the same way for my long
portfolio: I want these stocks to increase in value as much
as possible.
So in the end the question is about minimizing variance in
general [or not]. At the moment, this is what I'm trying to
do (also I agree with your concerns on Markowitz-like
frameworks and will keep your suggestions in mind).
G
Pour
: <r-sig-finance at stat.math.ethz.ch>
cc :
Objet
: Re: [R-SIG-Finance] solve.QP (for portfolio optimization)
"Christian Prinoth"
<Christian.Prinoth at epsilonsgr.it>
Envoy? par :
r-sig-finance-bounces at stat.math.ethz.
ch
10/01/2007 14:35
If the goal is to build some kind of market neutral position,
one could also take a 2-step approach:
1) build an optimized long portfolio that maximizes some
score\expected return\whatever
2) build a short portfolio that minimizes that same score,
while constraining risk exposure (however defined) to be
similar to that of the long portfolio.
This way it is easier to specify leverage, number of positions etc.
Christian Prinoth
cp at epsilonsgr.it
+39-0288102355
-----Original Message-----
From: r-sig-finance-bounces at stat.math.ethz.ch
[mailto:r-sig-finance-bounces at stat.math.ethz.ch] On Behalf
Peterson
Sent: Wednesday, 10 January, 2007 14:06
To: r-sig-finance at stat.math.ethz.ch
Subject: Re: [R-SIG-Finance] solve.QP (for portfolio optimization)
On Wednesday 10 January 2007 06:26,
guillaume.nicoulaud at halbis.com wrote:
--- Brian wrote ---
"You may find, as many others in the optimization
that the short portfolio requires a different optimization
I did... and it actually doesn't work! That's why I would like to
optimize the whole portfolio instead of doing this
and shorts (in a Markowitz-like framework *for now*).
Markowitz style optimization will try to minimize variance
entire portfolio. You *want* the short portfolio to
as much as possible.
While it should be possible to constrain individual
on the short portfolio, I haven't worked with the solve.QP function
constraints in enough detail to give you any pointers there, and I
don't think a minimum variance portfolio is really what you want.
Perhaps you can be a little more specific on the problems
trying to optimize the long and short portfolios separately?
I'll give a couple examples of approaches that could work well for
your short portfolio (your exact circumstances will vary
instruments you're constructing a portfolio over, of
short portfolio, you have previously made some forecast that the
instruments in the short portfolio will decline in value.
make some decision about how much to short, from the limits
on total short positions in your portfolio. One method of
much to short is based on your confidence in your price
confidence equals a larger short position. Another method
some other appropriate measure of risk, like downside deviation or
average
drawdown: larger [downside risk measure] equals larger