Here is my latest blog update:
http://blog.alphascanner.com/?p=665
Billy
Here is my latest blog update:
http://blog.alphascanner.com/?p=665
Billy
Billy,
What is your experience with ATR around market bottoms and tops? I've read in many places that markets tend to become more volatile when the trend is changing and am wondering if this is actually true. Have you or someone you know done any actual studies on this topic?
Thanks.
Adam,
ATR and other volatility indicators tend to contract into consolidations. Many top traders like John Carter or Hubert Senters have developed "volatilty squeeze" systems. If a top or bottom is resolving into a consolidation with abating volatility, volatility will normally surge on any breakout from the consolidation.
My own proprietary risk management system, also used by the robots, tries to take advantage of this phenomenon with optimal trailing stops linked to contracting or expanding volatility. In uptrends, ATR % will tend to contract most of the time and the trailing stop will rise exponentially faster over time. Because starts of down trends are much more violent and volatile, the trailing stop is quickly hit on the reversal and you can capture most of the profits this way.
Billy
Adam,
ATR and other volatility indicators tend to contract into consolidations. Many top traders like John Carter or Hubert Senters have developed "volatilty squeeze" systems. If a top or bottom is resolving into a consolidation with abating volatility, volatility will normally surge on any breakout from the consolidation.
My own proprietary risk management system, also used by the robots, tries to take advantage of this phenomenon with optimal trailing stops linked to contracting or expanding volatility. In uptrends, ATR % will tend to contract most of the time and the trailing stop will rise exponentially faster over time. Because starts of down trends are much more violent and volatile, the trailing stop is quickly hit on the reversal and you can capture most of the profits this way.
Billy
Billy,
When you say the trailing stop will rise exponentially faster over time, why is that? I assume it has to do with contracting volatility but not sure how the math works to cause the stop to move this way.
Also, I'm trying to figure out how to use ATR in such a fashion that it can "target" drawdown percentages. In other words, if I'm not comfortable with the historical intra-trade drawdown of an instrument, is there a way to manipulate ATR to maximize return while keeping drawdowns to the stated percentage.
I confess I'm relatively new to the ATR concept.
Adam,
1) The trailing stop is a function of ATR% deducted from the highest closing price. The faster the ATR% decreases, the faster the trailing stop will get close to the trading price.
2) It is foolish to try to use ATR for targeting fixed drawdown percentages. If you can't stand more than X% drawdowns, just use a fixed X% percentage stop. But your performance will decrease drastically unless you sell often on targets. But even such a strategy with optimal targets will underperform significantly risk-adjusted returns from a trailing ATR% stop.
Billy
Last edited by Billy; 07-06-2011 at 12:25 AM.
Very useful information Billy which makes me think of J. Welles Wilder's parabolic system (Stop and Reverse or SAR) presumably just on the long side.
I presume that the same exponential approach is valid when one is short? Yes or No
or the model is different for a short trade because your research has shown a different optimal stop strategy when short.
Pierre
The term "X% drawdown with a fixed stop" is really a MAE (max adverse excursion) for a single trade, which is not the same as account drawdown on open equity spanning consecutive trades. Without extensive backtesting, I believe it's impossible to arrive at drawdown estimates for any given stop strategy.
Trader D