|
PureBytes Links
Trading Reference Links
|
The problem is that the most meaningful performance statistics are those
calculated on a portfolio basis. For portfolio-based back-testing I use
Trade-Sim and export the monthly portfolio equity into Excel where I can do
statistical analysis. From that you can then get data you can compare
against published results from hedge funds or CTAs.
Shown below is a set of results from my Excel spreadsheet. However, this
process is anything but automatic.
Average Annual Return 28.52% Best Monthly Return
14.07% Beta vs S&P -0.10
Annual SD 11.02% Worst Monthly Return
-7.00% Alpha vs S&P 0.01
Skewness (monthly) 0.95 Best Annual Return
43.85% Correl Coef -0.16
Kurtosis 3.64 Worst Annual Return
2.71% R squared 0.03
Annual Sharpe Ratio 1.18 % Winning Months
66.32%
Annual Sortino Ratio 1.89 Worst Drawdown
6.05%
If you want to think about this in more detail, look at Black "Managing a
Hedge Fund" and McCrary "How to create and manage a Hedge Fund". They both
have chapters on performance metrics and Black includes a lot of comparable
data on different investment styles.
How do others do this?
Andrew
-----Original Message-----
From: kut2k2 [mailto:no_reply@xxxxxxxxxxxxxxx]
Sent: Monday, January 10, 2005 10:43 PM
To: equismetastock@xxxxxxxxxxxxxxx
Subject: Re: [EquisMetaStock Group] Should I follow indicators?
--- In equismetastock@xxxxxxxxxxxxxxx, "Andrew Tomlinson"
<andrew_tomlinson@xxxx> wrote:
>
> Interesting. I come out somewhere in the middle. I think you need
to
know a
> fair amount in order to have the courage to stay the course. If
you've given
> yourself a good grounding in TA, then it's harder for some
self-appointed
> guru to take you for an expensive ride based on his or her
interpretation of
> some basic indicators that you should have known about anyway. If
you've
> learnt enough to do some back-testing of your simple idea then
you'll have
> the fortitude to follow your system through drawdowns.
>
> So at least do some basic homework (e.g. Murphy "Technical
Analysis...")
> before you start, and learn to do some backtesting.
>
> Oh, and be careful re Van Tharp. His mathematical rigor leaves a
lot
to be
> desired, although many of the ideas that he summarizes from good
traders are
> ok. And don't expect to be able to replicate any of his numbers.
>
> Andrew
I pretty much agree. If trading was really easy, we'd all be
millionaires. OTOH you don't need to read a 100 books to get a decent
strategy -- not a perfect strategy but at least a decent one. I think
the one thing we all agree on is that you have to verify FOR YOURSELF
what works.
To that end, which performance evaluators do you like the best, Andrew
(and others)? I'm at that point where it seems as if there's a million
different performance schemes, almost as bad as the number of
different indicators, and some guidance here on what works for
evaluating a system and what doesn't would be much appreciated.
Good trading,
kut2k2
>
>
> -----Original Message-----
> From: Pam and Ken Johnston [mailto:milocat@x...]
> Sent: Monday, January 10, 2005 7:14 PM
> To: equismetastock@xxxxxxxxxxxxxxx
> Subject: Re: [EquisMetaStock Group] Should I follow indicators?
>
>
> I hate to contradict Superfragalist because its obvious from his
posts that
> he knows that about which he speaks but I have to offer a
completely
> different opinion. Trading is made unnecessarily complicated by
most
> people. I remember reading a book about top traders and one of the
> interviewees made a refreshingly different observation about
trading,
> essentially he said that the good news was that finding a good
method to
> trade that produced excellent results over time wasn't very
difficult. The
> bad news was that having the discipline to stick to any given
method
was way
>
> beyond most people.
> I would forget about 100 different books and psychological
consultants.
> Find a simple method that makes sense to you(otherwise you will
abandon it)
> and resolve to stick with it. Astrocycles-time extraction-fibo
squares etc
> are likely no better than moving averages-the tough part is picking
a method
>
> and staying the course.
> I doubt that there is any facet of human existence where the KISS
principle
> is more applicable than trading.
> ----- Original Message -----
> From: "superfragalist" <no_reply@xxxxxxxxxxxxxxx>
> To: <equismetastock@xxxxxxxxxxxxxxx>
> Sent: Monday, January 10, 2005 9:44 AM
> Subject: Re: [EquisMetaStock Group] Should I follow indicators?
>
>
> >
> >
> >
> > Yeah, I have a suggestion. You should read somewhere around a 100
> > trading and systems development books, and then if you plow
another
> > few thousand hours into understanding what you've read,
experimenting
> > and figuring out how to trade with what you've come up with, then
> > you'll be close, really close.
> >
> > There's nothing anyone on here is going to post that will help
you
> > more than the above mentioned paragraph.
> >
> > If there was one good indicator--or trading method--we'd all be
using
> > it. We ain't. Indicators fit people and personalities. It's all
in
> > what you can use, and no one on here, or anywhere else, can
define
> > that for you.
> >
> > Go to amazon buy every book they have on trading and sytems
> > development and you'll be off to a relatively mild start, but
> > generally in the right direction.
> >
> > Read Van Tharp's Trade Your Way to Financial Freedom first. That
will
> > get rid of a lot of your misconceptions like entries are
important.
> > Read the other 100 books next, and then read Van Tharp's again.
By
> > that time you might have a prayer of getting it.
> >
> > If you want easy--go bowling!
> >
> > Sorry, but them's the realities.
> >
Yahoo! Groups Links
------------------------ Yahoo! Groups Sponsor --------------------~-->
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/Tcy2bD/SOnJAA/cosFAA/BefplB/TM
--------------------------------------------------------------------~->
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/equismetastock/
<*> To unsubscribe from this group, send an email to:
equismetastock-unsubscribe@xxxxxxxxxxxxxxx
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|