Search results

For a thoroughly backtested system, with enough sample size (say 1000+ trades), your simplest method would be to double the historical max drawdown, and make this your system stop. To go from 1 contract to 2 contracts, your account must grow by at least the amount of that system stop + margin...

If you have enough historical data (say 10 years), backtest the basic mean reversion system & record for each entry as much contextual information as you can think of. Then do statistical analysis, and see if you can come-up with well identified contexts where there is a clear edge in either...

I would judge any and all platforms on the level of support provided for live, 100% hands-off automated trading. To me, the most important thing is the ability for a live traded strategy to recover from a loss of connectivity, and/or a plain restart, without losing track of its positions and...

What worsen matters, markets, while not random, are (of course) only probabilistic (and I certainly agree w/ the low signal/noise ratio, which further aggravate things, by both "hiding" true signals in the noise, as well as "making up" false signals from noise).

Most just lose from bid/ask spread & commissions. They would lose about the exact same doing the opposite.

I have a system which generates ~ 10,000 trades in 6 years on 1 market, and I am still suspecting overfit ... ... if your EOD system trades stocks, then you should be able to reduce the risk of overfit by backtesting over a reasonably large set of stocks. Going to 20 years only doubles the...

IMO ... 2 years of historical data isn't enough to be representative of the various modes a market can be in. You could have 5000 trades in a 2 years (intraday) backtest, and still have a large exposure to the next regime change. If your intraday strategy has good results on 2 years, you...

Take a look at Robert Pardo 2nd edition : The Evaluation & Optimization of Trading Strategies. A decent read for someone just starting with ATS.

It is pretty clear this is motivated by predator firms ... ... easy to see how this will be used to conveniently "stop" the market somewhere with a large resting sell order, accumulate quietly, then go long some at MKT which will auto-cancel the large resting sell order and screw all sellers...

I am not trying to sell you anything and I don't have to prove you anything either. I offered you my best advice, feel free to ignore it, as I am ignoring you moving forward. May-be, take an elementary-grade Math course ... 26,000 / (6*250) = 17 trades / day on average. Not HFT in my...

How about that? (2007-2012)

I am in regular contact with NetTecture for about 9 months now, and while I respect him for what he does, I also know that he is not the one programming his strategies, he has a couple of guys doing that for him, and who knows how good they really are? Anyway, I am the one who pointed out to him...

Hftvol, you have very good arguments in favor of professional-grade platforms, however, you are missing the point ... most ETers are not professional, merely trying to get their feet wet, discover and learn, and there is no point for that purpose to spend the kind of money required by a...

I have a system which is always-in (long or short), and currently makes directional bets upon a specific timing signal. The direction of the bet results from statistical analysis of price action after said timing signal. So this system has no stop, it can only reverse a bad position at the...

Sorry I really don't see the humor in that. Given that there is 90% chances you won't get to a profitable strategy that can be traded live, this is saving you a mere $995 for Ninja, and you still get the benefit of their customer support. And if you hit that 10% where you actually can go live...

j2ee, - what's your discretionary trading experience? - what's your automated systems design experience? Your first (and largest) hurdle when starting to develop automated systems isn't going to be live execution, but simply getting to your first tradeable strategy. This is really the...

I suppose this is to store bars in RAM once historical data is read from file, for the purpose of chart display / data analysis / indicator or strategy processing. - Add Volume and Ticks to your Ohlc class (I would add a DateTime for the bar's close DateTime, too) (I would also keep track of...

This entire thread is plain stupid. I am not much impressed with most of the comments, and what they reveal from the people behind the keyboards. Surf, I hope you realize calling one trade once in a while is pointless. Even the best odds setup can turn into a loss. The business of trading...

To be honest, what anyone believes is of no interest whatsoever. You should backtest the different approaches that you identified, and come to your own conclusions, based on "facts".

Finding swing highs & lows is very easy when you start on that ... just define a swing high as an N-ring pivot, meaning it is the highest bar in a group of 2N+1 bars, with N bars stricly lower than the swing high on its left side, and N bars lower or equal to the swing high on the right side...