Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Occasionally the SuperTrend stop (trend?) indicator is completely contradicted by the price action that follows. Here's a particularly nasty example on TF today. Is there any configuring I can do in the parameters to increase the reliability of this?
Thanks, I'm pretty new to NT and nexusfi.com (formerly BMT) so I hope this is posted correctly.
Geoff
Can you help answer these questions from other members on NexusFi?
Indicators are not nasty. You can use them in different ways.
The SuperTrend is a trailing stop based on volatility. The SuperTrend has not been designed for entry timing, as it will only give delayed signals. However, it can also be used as a higher timeframe trendfilter. Whether it is suited for following price action or not depends on several conditions.
(1) The SuperTrend works in a trending environment. Therefore it is important to know the probability that the market will be trending when it is used.
(2) Depending on the settings it is important to have an understanding of the
- intraday/daily distribution of volatility
- intraday/daily distribution of returns
The main characteristics of those distributions can be shown, when applying a z-score indicator to the ranges and bar-to-bar returns.
3-sigma events for volatility on daily chart: 3/184 = 1.63%
2-sigma events for returns on daily chart: 16/184 = 8.70%
3-sigma events for volatility on 5-min chart: 5/184 = 2.72%
2-sigma events for returns on 5-min chart: 16/184 = 8.70%
Looking at the charts, there are two possible educated guesses:
-> the distribution of returns on a daily chart is assymetric (there are more 2-sigma events to the downside)
-> the distribution of volatility on an intraday chart has a higher number of 3-sigma events compared to a daily chart
Both educated guesses can be more or less confirmed, when performing similar tests with other instruments. However, currencies seem to behave in a different way, as the intraday distribution of 2-sigma events for returns was less frequent.
We can now draw a number of conclusions:
-> currencies are probably trending better than other instruments and more suited for any approach relying on the SuperTrend
-> intraday volatility for most of the futures contracts shows a higher number of outliers than expected, which make strategies basd on the SuperTrend more vulnerable
-> it is important to have an idea whether the current day will be a trending day (following news release) or not
I would further suggest to add a second higher timeframe SuperTrend. Thus any positions against the higher timeframe trend would not be allowed.
Attached is a sample chart. As you can see the losing trades have been mostly eliminated.
@Fat Tails, would you be so kind as to post the code for anaSuperTrendU11Dual? I have been working with two timeframes as you suggest and have recently "jumped into the pool" coding Ninjascript instead of relying on the Strategy Wizard. Examples like this dual-timeframe indicator would be very helpful in my learning process.
Thank you gdstuart for asking this question and FatTail for your detailed information (as always). Please could you provide a some more information on the higher frame filter? Are you referring to adding this filter on the same chart or another higher timeframe chart?
Also, please could you code the indicator for NT7 as well?
The anaSuperTrendDual is just built from two anaSuperTrends. You can also add two SuperTrends one by one and get the same result. Depending on the selected settings, one of the SuperTrends is the higher time frame filter and the other one is the smaller time frame filter. The composite trend is determined as follows
- both SuperTrends up -> composite trend up
- both SuperTrends down -> composite trend down
- SuperTrends opposed -> composite trend neutral
The only thing that the anaSuperTrendDual adds is the shading for the uptrend, downtrend and neutral areas.
I'll be looking forward to it. Currently I add the longer timeframe to the plot but hide it, and then add two instances of anaSuperTrendU11. It does the job and I'm now trying to incorporate it into a strategy. Interesting possibilities for automation across many timeframe combinations.