My performance is ? no idea

Discussions about the testing and simulation of mechanical trading systems using historical data and other methods. Trading Blox Customers should post Trading Blox specific questions in the Customer Support forum.
Post Reply
painless
Roundtable Fellow
Roundtable Fellow
Posts: 59
Joined: Tue Nov 14, 2006 12:52 pm

My performance is ? no idea

Post by painless » Fri Nov 17, 2006 8:47 am

I can find no reference to what anyone is using as an initial account balance. Without knowing this you can quote all sorts of statistics but it is apples and oranges unless we are all using the same initial amount.
Is there some secret number that I don't know about?
I have enjoyed reading the posts and I think there are a lot of experienced traders here. I would like to get some idea of how my work is progressing so can someone please enlighten me.

Thanks

LeapFrog
Roundtable Knight
Roundtable Knight
Posts: 695
Joined: Mon May 17, 2004 4:18 pm
Location: Boston, MA

Re: My performance is ? no idea

Post by LeapFrog » Fri Nov 17, 2006 9:05 am

painless wrote:I can find no reference to what anyone is using as an initial account balance. Without knowing this you can quote all sorts of statistics but it is apples and oranges unless we are all using the same initial amount.
Is there some secret number that I don't know about?
I have enjoyed reading the posts and I think there are a lot of experienced traders here. I would like to get some idea of how my work is progressing so can someone please enlighten me.

Thanks
I can understand your urge to know what dollars people are trading and wanting to relate that to your own situation. I would strongly encourage you though to resist this temptation. It really becomes irrelevant over time because everybody's situation is different. Different net worths, different personal debt situations, family obligations, fund sizes if we are managing money. Yes, bigger accounts can trade more products and more systems but the small trader can dart in and out quicker than the big boys.

What is more helpful, I think is measuring percentages and drawdowns. This you can use to compare your performance with against everyone. Look at the IASG database, for example. http://www.iasg.com/mainframe.asp?
Or look up specific fund prospectus and see their monthly performance percentage changes. I track such things and try to pick out commodity futures funds close to the way I trade and compare my results to those - in percentage terms.

Hope this helps.

painless
Roundtable Fellow
Roundtable Fellow
Posts: 59
Joined: Tue Nov 14, 2006 12:52 pm

Post by painless » Fri Nov 17, 2006 9:23 am

I only want to know how to relate my performance not how much money others have. That was a nice link you gave me but I am no wiser and I checked the FAQ and statistics definitions link but still no joy.
YTD return on what? I cannot measure percentages on anything without a nominated start capital.

painless
Roundtable Fellow
Roundtable Fellow
Posts: 59
Joined: Tue Nov 14, 2006 12:52 pm

Post by painless » Fri Nov 17, 2006 9:56 am

Ok my fault. I don't want to know your account balance. I need to know how much equity is allocated per whatever is being quoted.
So if someone says there drawdown is 30% on JY then it would have meaning if they quantified the starting capital per contract.
eg: exchange margin * 2
or some such
When you test your system using whatever it is you use then you must start with something otherwise it is not possible to calculate a percent return.

If traders are testing with the option of compoundie ie: adding contracts once there account reaches a certain point then that also affects any chance we have of making comparisons.

BARLI
Roundtable Knight
Roundtable Knight
Posts: 650
Joined: Sat Jan 17, 2004 6:01 pm
Location: USA

Post by BARLI » Fri Nov 17, 2006 10:21 am

I test using 300k as initial capital, my real account is way smaller than that. I chose 300k for portfolio simulations as some realistic number, since 1 mil$ is something not real to raise at this point of time

PaulZ
Roundtable Fellow
Roundtable Fellow
Posts: 95
Joined: Thu Oct 06, 2005 9:59 am
Location: Philadelphia Area

Equity Capital and Testing

Post by PaulZ » Fri Nov 17, 2006 11:03 am

Starting capital is very important in testing for a number of reasons:
1. smaller starting capital means smaller position sizes and fewer trades. For example, if you risk 2% of your equity on a trade, many trades will be ineligible due to their initial risk exceeding the 2% cosntraint.
2. using large starting capital (I use $1M in my testing) enables trading of more markets since fewer markets will be ineligible for trading due to exceeding the 2%. That enables me to see how well a system works across many markets, even though I don't have anywhere remotely close to $1M in my trading account. If testing with $1M starting capital gives favorable test results, it helps me to appreciate how universal the system may be across many markets.
3. lets say a system looks good over many markets, having tested with large starting capital. Now try testing with progressively smaller starting capital and see if the performance stats look similar. If you get good performance starting with $1M and your performance is considerably reduced starting with $100K, then that gives you lower confidence about what performance you will see with small starting capital. BTW, by "good performance" I'm referring to things like CAGR/MAR, NOT to profit. What gets me really excited is if I get good performance going all the way down to $20K. Then I have greater confidence that I can start successfully trading with smaller account size.

painless
Roundtable Fellow
Roundtable Fellow
Posts: 59
Joined: Tue Nov 14, 2006 12:52 pm

an example

Post by painless » Fri Nov 17, 2006 11:05 am

Ok let me give an example.
I am currently testing USDX futures. This is what is turning up so far without using any stops. Just trading off the tool.
From = 02/11/1987
To = 14/03/2000
One Contract only - no compounding
Net = $66140
Max DD = -$6185

So how can one calculate anything in % terms. It is impossible. A starting balance must be nominated.

painless
Roundtable Fellow
Roundtable Fellow
Posts: 59
Joined: Tue Nov 14, 2006 12:52 pm

Post by painless » Fri Nov 17, 2006 11:09 am

Gee I am just a minion. I started with $10000 because that was the minimum the broker would accept. A $200 drawdown won't get me very far.
javascript:emoticon(':lol:')

painless
Roundtable Fellow
Roundtable Fellow
Posts: 59
Joined: Tue Nov 14, 2006 12:52 pm

Post by painless » Sat Nov 18, 2006 12:24 pm

Now to answer my own question so I look completely stupid. I did not notice the margin/Equity Ratio on the iasg site ... oops!

so on my current USDX test which varies a little from the previous post.
Margin/Equity would be 8.5%
Drawdown is 41%
Annualised return is 39.5%

on the market I currently trade
Margin/equity is 16%
Drawdown is 33%
Annualised return is 118%

ok sorry for annoying you all and thanks for the help

painless
Roundtable Fellow
Roundtable Fellow
Posts: 59
Joined: Tue Nov 14, 2006 12:52 pm

Post by painless » Thu Nov 23, 2006 12:53 pm

Hi all,
Looks like I am still wrong, oh well. I took a good look at the figures on that IASG website. Assuming that the margin/equity ratio is maintained then the results that are shown are compounded. Whereas my margin/equity ratio when testing only one contract becomes ever smaller. Sorry for taking up all this forum space by not studying all the terms et al. I tend to live in my own world. What a nong!
So I would have to turn on compounding to make a valid comparison.
Since the best one over on the website 3 years is only 102% then I guess I am doing extremely well. So I must count my blessings and feel satisfied with that then.
Pity I am only trading with threepence but then maybe that is a blessing and I just don't know it. :)
cheers

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo » Fri Nov 24, 2006 10:15 am

If you want to compare the results of a private trader against the IASG results of a professional money manager who charges for her services, you may wish to calculate the answers to one or both of these questions:
  1. What would my results be, if I deducted the same fees as a pro would deduct?
  2. What would Ms. X (a professional whose results are listed on IASG) have gotten, if she traded her own money and didn't charge herself fees?
If you decide to use Excel to calculate the answers, #1 is a little bit easier than #2. Professionals charge "two and twenty", two percent annual management fee, and twenty percent of any new profits above the previous high water mark.

Set up your spreadsheet with your monthly equity values. First, deduct (1/12 of 2%) which is one month's worth of management fees. Then compare equity against the previous high water mark. If greater, deduct 20% of the excess (this excess is called "net new profits", and 20% of it goes to the manager). Presto you've got the monthly equity after fees and incentives. Do this for every month and when you're all done you'll have a new equity curve, the "including fees and incentives" equity curve. Now calculate the max drawdown%, duration of longest drawdown, CAGR, Sharpe, MAR, Sortino, Lake, Semideviation, RRR ratios on this "including fees and incentives" equity curve. Compare against Ms. X.

Calculation #2 is similar except you do it "in reverse" and the easiest way is an iterative "shrink the interval" loop in a standalone computer program which you can write in Delphi, Python, TCL, Java, BASIC, C++, or any other language your heart desires.

painless
Roundtable Fellow
Roundtable Fellow
Posts: 59
Joined: Tue Nov 14, 2006 12:52 pm

Post by painless » Fri Nov 24, 2006 11:11 am

thanks sluggo,
I think I get that. I did a few cagr numbers on my output and I can aproximate my results enough to make a quick and dirty comparison. I don't have to be the best in the world so long as I making a bit. When it is all said and done I look at the buys and sells on my chart and if I feel happy about what I see then that is a good sign. Secondly after trying a lot of different ratios and stats I have found that my simple flat %pa on one contract is a reasonable guide when making comparisons between my own tests. Now I think I have a handle on what the pros are doing in comparison.
Thanks again
Blair

LeapFrog
Roundtable Knight
Roundtable Knight
Posts: 695
Joined: Mon May 17, 2004 4:18 pm
Location: Boston, MA

Post by LeapFrog » Sat Nov 25, 2006 10:07 am

sluggo wrote:If you want to compare the results of a private trader against the IASG results of a professional money manager who charges for her services, you may wish to calculate the answers to one or both of these questions:
  1. What would my results be, if I deducted the same fees as a pro would deduct?
  2. What would Ms. X (a professional whose results are listed on IASG) have gotten, if she traded her own money and didn't charge herself fees?
If you decide to use Excel to calculate the answers, #1 is a little bit easier than #2. Professionals charge "two and twenty", two percent annual management fee, and twenty percent of any new profits above the previous high water mark.

Set up your spreadsheet with your monthly equity values. First, deduct (1/12 of 2%) which is one month's worth of management fees. Then compare equity against the previous high water mark. If greater, deduct 20% of the excess (this excess is called "net new profits", and 20% of it goes to the manager). Presto you've got the monthly equity after fees and incentives. Do this for every month and when you're all done you'll have a new equity curve, the "including fees and incentives" equity curve. Now calculate the max drawdown%, duration of longest drawdown, CAGR, Sharpe, MAR, Sortino, Lake, Semideviation, RRR ratios on this "including fees and incentives" equity curve. Compare against Ms. X.

Calculation #2 is similar except you do it "in reverse" and the easiest way is an iterative "shrink the interval" loop in a standalone computer program which you can write in Delphi, Python, TCL, Java, BASIC, C++, or any other language your heart desires.
I've done this re-engineering of some public commodity funds also. What I find revealing is how long the dry spells can be with no performance fee - years can easily go by if preceeded by a particularly good year. If the fund then goes into a 50% type drawdown, I can see why the motivation might be to shut down the fund. Especially a small one of say just a few million.

My point would be that the appeal of setting up one's own hedge fund versus staying independent may not be that appealing unless some serious momentum could be established out of the gate - like starting with $50 to $100 Million ("It could happen!"). Alternatively, if intending to manage a small (<$50Million) fund, make sure it is with just a handfull of very serious investors with long lock up periods and keep startup costs and overheads to a minimum (live off the fees).

BARLI
Roundtable Knight
Roundtable Knight
Posts: 650
Joined: Sat Jan 17, 2004 6:01 pm
Location: USA

Post by BARLI » Sat Nov 25, 2006 5:54 pm

LeapFrog, did you ever deal with money raising? from what I understood from others guys here its not a pleasant thing to do

AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner » Sun Nov 26, 2006 3:50 am

Barli

I have long thought about this and its "unpleasantness". It is not just that trend following or futures can be a hard sell. It is not just that a start up CTA with an 18 month track record and 10 cents under management will find it hard to get in the front door - hence the chicken and egg situation.

What it really boils down to (in my view) is character and determination. Some people are natural salesmen. They are gregarious, they enjoy meetings and lunches and dinners and operas with friends and contacts and potential clients. For these guys, while the road will doubtless be a long one and they will need to put in a great deal of effort, money raising is not going to be "unpleasant". It will flow naturally from what they do well.

Others are "anoraks". They enjoy fiddling with the research, they enjoy devising and testing new systems, they revel in figures and spreadsheets and really getting to know their subject inside out.

A lucky few may be both solid anoraks and excellent salesmen but I suspect the two are often mutually exclusive. Maybe the most successful money managers are both technically good and extravert showmen?

RedRock
Roundtable Knight
Roundtable Knight
Posts: 939
Joined: Fri Jan 30, 2004 3:54 pm
Location: Chicago

Post by RedRock » Sun Nov 26, 2006 1:51 pm

AFJ Garner wrote:Barli

IA lucky few may be both solid anoraks and excellent salesmen but I suspect the two are often mutually exclusive. Maybe the most successful money managers are both technically good and extravert showmen?
Seems like an extension of right vs left brain dominance. To have both is less common.

Some can design a structure, some can build it. A few, can do both.

BARLI
Roundtable Knight
Roundtable Knight
Posts: 650
Joined: Sat Jan 17, 2004 6:01 pm
Location: USA

Post by BARLI » Sun Nov 26, 2006 2:16 pm

AFJ Garner, could you share some of your own experience? For instance: what is usually the drawdown when customers start calling and crying?

AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner » Sun Nov 26, 2006 2:27 pm

Barli

I do not manage money for others - I have run a proprietory trading book since leaving the world of investment banking in 1992, sometimes with partners. I did launch a Cayman Islands fund at one stage, with partners, but their idea of fund management did not accord with mine and we parted company. Others are far better qualified therefore to answer your question

A

PS - my vote is with the anoraks; I prefer analysis to sales. I also loathe corpoarate bullshit!
Last edited by AFJ Garner on Sun Nov 26, 2006 2:37 pm, edited 1 time in total.

BARLI
Roundtable Knight
Roundtable Knight
Posts: 650
Joined: Sat Jan 17, 2004 6:01 pm
Location: USA

Post by BARLI » Sun Nov 26, 2006 2:32 pm

ok then maybe someone else will share :)

Post Reply