0
$\begingroup$

Let's say I have a simple strategy that involves going long on a stock whenever it rises above its 50 day moving average. As most are probably aware, this type of indicator works well in capturing sustained upward price movements and does equally well avoiding long downward price movements, but can get killed when price action is choppy.

So a simple question - can threshold-type indicators be modified to avoid over-trading in times of choppiness while still staying true to the goal of being on the right side of the trade for extended price movements?

$\endgroup$
2
$\begingroup$

If A is true then B is false and if B is true then A is false. You're asking for something that works when A is true and when B is true. It doesn't exist.

Trend following indicators work well in trending markets. Well, sort of. They're late in and late out. They're not effective in sideways markets because they will generate false signals and whipsaws.

Technical analysis indicators provide information like support and resistance, current trend, and current momentum but they are merely a reflection of past price and volume. They predict absolutely nothing going forward. It's like looking in the rear view mirror and expecting that to tell you where you are going.

Any trade that you take based on such analysis is based on the HOPE that whatever trend or momentum you have identified will continue. Indicators predict absolutely nothing going forward so there's no way to know if the ensuing move after a reversal is going to be choppy or will be the beginning of a new trend.

$\endgroup$
4
  • $\begingroup$ I guess I should have offered a suggestion or two in the question. Just a couple of examples off of the top of my head - 1) using a MA that changes its lookback period based on some metric, 2) some kind of +/- buffer that captures a time element and a magnitude element (e.g., price trickling below MA is treated different than price crashing through it). Again it was open-ended question. I'm sure people out there have found ways to improve on basic indicators. $\endgroup$ – SuperCodeBrah Dec 31 '20 at 21:02
  • $\begingroup$ I should also mention that I just used MA as an example, but the indicator I'm using is a quantitative volatility indicator but not really a "technical" indicator like MA, RSI, etc. It's designed to identify periods of high and low risk in the market to scale leverage but appears to suffer from a similar problem in choppy markets. $\endgroup$ – SuperCodeBrah Dec 31 '20 at 21:05
  • $\begingroup$ Anything that changes the lookback period of a MA is akin to curve fitting, unless you're suggesting parameter adjustment based on current volatility. But either way, nothing can predict forward movement. The idea of of a magnitude element (e.g., price trickling below MA being treated different than price crashing through it) is somewhat akin to a stop loss order. Nothing happens until X is reached and by the time you reach X, you have given back X. Neither the stop loss nor any indicator can determine how fast X is going to be or even will be reached $\endgroup$ – Bob Baerker Dec 31 '20 at 22:12
  • $\begingroup$ Yes, volatility or maybe Hurst exponent (although I've never figured out how to make this into an indicator) could be one such metric to adjust the lookback period, but whatever the method, I'm talking about something formulaic. If a strategy is simple enough and rule-based, I would think curve fitting would be minimal. To the point about reaching X, the idea would be that your X is dynamic and placed such that the likelihood of a continuation is maximized when reached. But this is more specific than I was intending - I was just looking for generic improvements to threshold-based strategies. $\endgroup$ – SuperCodeBrah Dec 31 '20 at 23:46
1
$\begingroup$

You face a trade-off between 'smoothness' and lag. There are moving averages designed to address that (e.g. see John Ehlers work) but it's not clear they will offer a meaningful improvement in performance - though that's something you can check for yourself.

$\endgroup$
1
  • $\begingroup$ One way address the issue of smoothness and lag is double smoothed moving averages. I think that I first read about them in the mid 1990's in Technical Analysis of Stocks & Commodities. In addition to that, Bill Blau developed a number of indicators that emploted double smoothing and wrote a number of articles for TASC as well as a book Momentum, Direction, and Divergence: Applying the Latest Momentum Indicators for Technical Analysis. $\endgroup$ – Bob Baerker Dec 31 '20 at 22:05
0
$\begingroup$

There are quite a few indicators that attempt to mitigate whipsaw. An alternate approach is to simply avoid trading in choppy price trends. For example, try pairing your strategy with the use of a threshold level on the Choppiness Index by E.W. Dreiss.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.