Doing a statistically-based analysis (i.e. calculating means, labelling outliers) on such a small sample size is of almost no use and is likely more misleading than anything. With this small amount of data, there is no basis whatsoever to label the first day as an "outlier", since we don't have nearly enough data to reliably characterize the statistics of the daily returns. Frosty, it has been said before, but bears repeating, if your backtesting results show a longterm profit and you trust them (both of which seem to be the case) and you've done the proper money management/risk calculations to ensure that you can handle the worst times you project, you just need to hang in there. One suggestion, if it hasn't been said before, it seems that much of your anxiety comes from the nature of your algorithm's returns, you said you expect it to have a few big winners, spread between many smaller losers, correct? It's natural that such a "herky-jerky" equity curve would cause nervousness. If you can't handle the ups and downs, try to modify the algorithm to achieve a smoother curve (a lot to ask, I know), but otherwise just have faith that IN THE LONG RUN (and I emphasize that because you are nowhere near approaching the long run, yet), things will balance out. Good luck and thanks for your candidness, I'm watching with much interest.
Frost ain't new to this, and neither am I. This is the same "system" (or a very similar one) Frost has been running off-and-on again for months with live cash, it's just re-emerging in this thread. I'm not basing my observations on 11 data points, but rather on a conversation which has been going on for months now regarding Frost and his trading Bots. Hey, it's not my money ... he's been given information by at least three or four traders that makes a lot sense to me, if it doesn't make sense to him, fine. Yes of course, you're curious to see how Frost does, and since it's his money on the line, it's no loss to the you one way or the other. But how would you feel when the Risk:Reward ratio inverts and drops to 2:1 and it were your dollars-at risk? Yep, thought so. JJ
Hello, I am an automatic trader from France watching your thread for two weeks. Damn! You are just experiencing the beginning of a drawdown, it's not even the fourth of what a real drawdown could be!What did you make on the overall? Down 500 $ or something like that... Don't plan to change anything so soon.. I 've been running systems for 3 years now, quite successfully( 35 % a year ... ) on futures, and when I start a new system , I don't stop it until I break the double of the Initial Maximum Intraday Drawdown from backtests. It's really tough, but It's the deal. If you respect the rules you have imagined at the beginning, be sure you will be rewarded. Since I trade 3 different systems, it's easier for me, 'cause my systems are compensating each others. If you're confident in what you did, don't stop it so soon... On the other side, I would say it's that I would never trust in a system that makes 3-4 trades per day unless it has some real strong " fundamental" causes. Like you are using the increasing noise, or doing arbitrage, or taking your signals on something that ER2 follows... not relying on stupid indicators, especially on ER2 ... Hope the best...
So if you're basing the analysis you previously presented on more than just the data your presented in your recent data copy/paste post, I'd be interested to see how you define what an outlier is, because to define an outlier, you need to have a good grasp on the statistics of the data. For example, what distribution do the daily wins/losses form? You presented an analysis, which appears incomplete, and I commented, but now you seem miffed that I didn't somehow know that you were basing your analysis on more than what you posted?
Like I said in the original post, I wouldn't be a big fan of the system to begin with because of its equity curve, I'd prefer something quite a bit smoother.
There have been lots of good advice flowing through this thread recently, and I would like to take a moment to thank everyone for their contributions and help thus far. I also appreciate everyone's interest in this thread and hope it continues to be constructive as it has been. I am happy with my switch to method 2 today. As I said earlier today after carefully examining how the methods trade I feel very confident that method 2 is the better method. It helps keep me out of the market more often and gets more bang for its buck when it does initiate a trade. The bot recently entered its first trade of the day a little while ago and is currently up +$357.... hopefully it remains possitive and gives me a much needed green day...
Update bot just closed out of its short trade for a profit of around $345 or so.... since this new method trades VERY few times per day.. this may be the only trade it makes today..... That is one of the bennefits of this new strategy..time in the market is MUCH lower than that of the origional method... which is good for my nerves if nothing else.. only several months will tell me if it was a good decision to swap methods.. but basically its almost identical to the first method just has one extra filter based on market internals.. so its not like I am trading a COMPLETLY new strategy...
I posted one last year around (December 2005). At that time it was "Mechanical Trading", not 100% fully automated. It was the ancestor of the ATS I am trading now. http://www.elitetrader.com/vb/showthread.php?s=&threadid=58841 I can post one for 2006 with some explanations. January 2006 is when I implemented the fully ATS, but I don't want to clutter Frost thread.
This is an equity curve of a strategy I developed that trades currencies on hourly bars. One of my favorite instruments is Euro-FX (6E), and I have been trading this strat for a little over 18 months. The backtest is over a 2.5 year span. The equity curve is normalized to an "average" position size of 4 contracts. There is a minimum required position size for this strategy. Profit Factor is a little more than 1.5, with minor drawdowns. RoughTrader