Manage Expectations, Adjust Leverage, And Manage Risk In Your Own Portfolio
Estimating Future Drawdowns
by Tushar S. Chande
You're going to get hit by a big one sooner or later, but how big is it going to be?
There are three unknowns about the future
performance of any trading system: the expected returns, the duration of
drawdowns, and the depth of drawdowns. Of the three, the depth of future
drawdowns is of the greatest interest because it threatens your survival
as a trader.
Drawdowns are retracements in equity from previous equity highs - in short, losing periods. Of course, the exact depth of a particular future drawdown for any trading system or trading manager is unknown, since many random factors can affect the outcome. However, even reasonable estimates of the worst-case future drawdown would be enormously useful for planning purposes.
First, accurate estimates prepare you for the worst that could occur, without the implication that the system has stopped working. Such estimates give you an important psychological edge: You are more likely to stick with the system or trading program through future drawdowns if its magnitude has been estimated in advance. You can resist the temptation to abandon a trading system or manager too soon. In the long run, abandoning a succession of strategies near the lows of their drawdown periods will significantly reduce your returns.
Second, if the depth of the expected drawdown is greater than you are willing to accept, then you can adjust your leverage - that is, volatility - to reduce expected future drawdowns to more comfortable levels.
Third, if you cannot adjust your leverage - as in the case of mutual fund investments - then you can look at other investments.
Fourth, from a system-design point of view, you can assess the effect of money management algorithms on future drawdowns and make adjustments to limit drawdowns to comfortable levels.
Fifth, accurate estimates can help you compare investment alternatives on an apples-to-apples basis. Presumably, your threshold of pain is relatively constant across different investment alternatives. With an estimate of the drawdown depth, you can ask how much risk-adjusted return can be expected from alternative investment strategies if the drawdown is limited to the same level for each strategy. Since returns and risk are intertwined, superior risk-adjusted performance can, presumably, be suitably rewarded.
So, accurate estimates of future drawdowns are of great interest to traders, investors, and money managers alike. One word of caution: Estimates of future drawdowns can be exceeded in practice. Thus, there is always a chance, however small, that the severity of future drawdowns can exceed your worst-case estimates.
ESTIMATES
Use monthly data in your calculations, since monthly returns are easily available for most systems and trading managers. (These ideas can also be applied with daily data.) The first step is to calculate the standard deviation of monthly returns, which is denoted as sM (M for monthly). The formula for the standard deviation of a sample (of returns, in this case) of size N (with N months of data) is given below:
...Continued in the July 2001 issue of Technical Analysis of STOCKS & COMMODITIES
Tushar Chande is a commodity trading advisor who has published Beyond Technical Analysis and was the coauthor of The New Technical Trader. A prominent innovator in technical analysis, his indicators such as Vidya, Cmo, and Aroon are included in many commercial technical analysis packages. He can be reached at golvcm@attglobal.net.
Excerpted from an article originally published in the July 2001 issue of Technical Analysis of STOCKS & COMMODITIES magazine. All rights reserved. © Copyright 2001, Technical Analysis, Inc.
Return to July 2001 Contents