My system results... what do you think?

what do you think?

  • The results are Ok/Good, I belive that you just might have a winner there

    Votes: 6 14.0%

  • There is no way you can achive these results for real.

    Votes: 14 32.6%

  • they seem OK...

    Votes: 7 16.3%

  • I've done/seen better...

    Votes: 8 18.6%

  • I don't know

    Votes: 8 18.6%
  • Total voters
    43
Quote from Avalanche:

Your max DD is over a quarter of a million dollars. Can you handle that.


Cheers.
More...

This where all these systems (based on historical data) fail miserably.

Who in there right mind could handle a drawdown like that based upon past performance?

NOBODY !

Why ?

Because they can not predict the future.


Watch out ET I see another vendor on the horizon.

That's what all these guys do!
 
Quote from Avalanche:

Your max DD is over a quarter of a million dollars. Can you handle that.

Also you have percent equity set to 15? Does that mean your risking 15% of our equity per trade, how do you have the trade size set. There should be something in there that tells you the avg or median size of the position both at the beg of the run at towards the end.

Stuff that scales up as you accumulate profits, you have to be real careful with.

Also just curious why do you think the equity curve goes vertical in 2003 most pullback stuff that is tested and includes 98-2000 kills it those years then gets hit hard in 01-02 then.....well you know, greatly out performs in the trendy mkts that lend themselves to pullback systems.

Thanks for posting and taking questions.

Cheers.
More...

Avalanche,

that drawdown translates to less than 30% of capital traded at that time, and for me that is acceptable.
(please take a look at the equity curve of the second system results I uploaded expressed in logarithmic scale, in which rise or decline is expressed in percentages...you will see that it seems much normal when viewed in relative terms).

No %Equity is not set to 15%, position sizing is done by a simuscript ( wld code for position sizing).
But to answer your question, the script does trade %equity, but different size depending on the signal (long or short). I defined it based on portfolio size and maximum number of open trades at one time in the past...I wanted to be able to take more than 90% of signaled trades because the equity curve gets erratic if I take too little of the trades that are signaled.

Equity goes vertical in 2003 only because it was expessed in absolute terms, in relative terms (% return) remains more or less steady.
Equity curve in in 2003 behaves the same as in 1999, just the rise is a little less steep in 2003.

Notice that in 2000-2002 exposure (light green) significantly drops( there is more dark green, which is cash on account) as long strategy is put on hold by a system itself (it kinda protects itself from unfavorable market conditions), and that most of the exposure comes from short trades(short strategy is based on totally different concept than the long). That is why the equity doesn't decline like in the other pullback buyers.

P.S. I uploaded the equity from second sys results rather than first because there is no that freak drawdown from short trades in 2001 so it is more representative of the points above, in all other respects is almost the same as the other one, I you want I can upload it too...
 

Attachments

Quote from Samson77:

This where all these systems (based on historical data) fail miserably.

Who in there right mind could handle a drawdown like that based upon past performance?

NOBODY !

Why ?

Because they can not predict the future.


Watch out ET I see another vendor on the horizon.

That's what all these guys do!
More...

Samson,

be rest assured I am not selling anything.

I just wanted some opinions because I'm planning to begin trading it next week...
 
Great post Morgan; good, useful info there.

thanks,

m

Quote from MorganSys:

Hello,

There are a few issues that you should consider adressing before implementing this system.

1. It sounds like the system does not act with the same rules long as short. A lack of symmetry in the rules allows you greater ability to curve-fit results. Hence, the short results (especially) are suspect.
2. Your data set is over 7 years long. Since you have more long trades than short trades, survivorship bias may be an issue (although it is unlikely change results to the full extent of the gains)
3. You have fixed percentages of the portfolio invested in positions. In a time series test, this can increase your ability to curve-fit results due to the varying leverage.
4. Your tests on other stock groups may not be completely independent, and will certainly not be if your system uses any market based variables or requires extreme price changes (which are to some extent systematically based).
5. Your maximum drawdowns were relatively recent. Has something changed?

I suggest the following tests:

1. Implement the long side rules for going long, and the inverse of the long side rules for shorting.
2. Implement the short side rules for going long, and the inverse of the short side rules for shorting.

---The market does not act symmetrically for all inefficiencies, so this may not be appropriate, depending on the logic of your system, use your judgement---

3. Hard to avoid survivorship without a database free of it. One way is to get the average beta of the stocks, the average return of the stocks, and find the market-adjusted returns of the stocks. The difference between this and zero is a very rough approximation of survivorship bias impact on a 100% invested portfolio.
4. Limit the number of positions to something that is usually smaller than the number of positions that the system wants to take (so you have a relatively constant fraction of your portfolio invested). To do this, you may want to test a very large number of stocks (the entire NYSE, entire NASDAQ, or the S&P 500). Alternatively: Position size = portfolio value / number of positions, and rebalance daily.
5. Look to see if similar strategies had poor results recently, and run tests every month or so to see if anything strange is going on. Also, test for a significant difference in results between the early years and the past year or so.
6. Run a test on only the past two years.
7. A chart of the results would be good.

You may PM me if you have any questions.

Best of luck,
Morgan
More...
 
maos,

Concerning your last equity curve,

It is quite troubling to see that buy and hold (the blue line) from March of 1999 to the the end of 2000 outperformed your system...

Or am I reading the blue line wrong..its hard to see the color code on my laptop after zoom?...

Michael B.
 
Quote from ElectricSavant:

maos,

Concerning your last equity curve,

It is quite troubleing to see that buy and hold (the blue line) from March of 1999 to the the end of 2000 outperformed your system...

Or am I reading the blue line wrong..its hard to see the color code on my laptop after zoom?...

Michael B.
More...

Yeah, you're reading it right. It really did outperform, but you can't have it all.
 
ah, that looks better; still some long periods between new equity highs. I can say from my own system trading, those can wear on you.

m

Quote from maos:

Avalanche,

that drawdown translates to less than 30% of capital traded at that time, and for me that is acceptable.
(please take a look at the equity curve of the second system results I uploaded expressed in logarithmic scale, in which rise or decline is expressed in percentages...you will see that it seems much normal when viewed in relative terms).

No %Equity is not set to 15%, position sizing is done by a simuscript ( wld code for position sizing).
But to answer your question, the script does trade %equity, but different size depending on the signal (long or short). I defined it based on portfolio size and maximum number of open trades at one time in the past...I wanted to be able to take more than 90% of signaled trades because the equity curve gets erratic if I take too little of the trades that are signaled.

Equity goes vertical in 2003 only because it was expessed in absolute terms, in relative terms (% return) remains more or less steady.
Equity curve in in 2003 behaves the same as in 1999, just the rise is a little less steep in 2003.

Notice that in 2000-2002 exposure (light green) significantly drops( there is more dark green, which is cash on account) as long strategy is put on hold by a system itself (it kinda protects itself from unfavorable market conditions), and that most of the exposure comes from short trades(short strategy is based on totally different concept than the long). That is why the equity doesn't decline like in the other pullback buyers.

P.S. I uploaded the equity from second sys results rather than first because there is no that freak drawdown from short trades in 2001 so it is more representative of the points above, in all other respects is almost the same as the other one, I you want I can upload it too...
More...
 
You may wish to consider:
1) Reducing slippage to 1c
- you can get VWAP for 1c as an institution (although this does have some slippage in it as well)
- as an individual with 100-300 shares, you should be able to enter positions with less than 1c slippage if you do smart order-entry, or use the opening auction on the NYSE for very low slippage.

2) Relaxing the constraints or slightly changing the logic so that you always invested in x% of stocks. If this crushes your system it is a bad sign. It is worrysome to me that your system frequently goes to cash, and that this is related to market movements. Macro timing is a very tricky thing since you have so few data points.

I also strongly suggest that for your entire development of future projects you test using a walk-forward algorithm for some time period that excludes the past year. After you are finished the system, then test it on the last year as a reality check.

Everything said, your test results remain fairly impressive. Best of luck with implementation.
 
In Wealth-Lab the Buy&Hold blue line represents buying the same amount of every stock which with a watchlist of over 100 stocks does not equate to to trading 10% equity. So I would not be concerned if you dont beat it.
 
You must log in or register to reply here.
Back
Top