The 50% level is the neutrality level. Think of it as a zero line, anything above it is a mostly bullish regime and anything below it is a bearish regime. It's the mid-point of the indicator.
The 55-bars ago comes from a technique I use to improve objectivity of the lookback periods I use. Basically, we can use, any number between 2 and infinity, which makes it tough to justify why we chose a certain number of periods. I use Fibonacci sequence to delegate this subjectivity. Hence, I may use 8, 13, 21, 34, 55, 89, etc... to at least not have to choose between a larger list which may include (10, 15, 20, 25, 30, 40, 60, 70, 80, 100, 200, 300, etc...).
Exit strategies are numerous and they're the ones that determine a good strat from a bad one. On this one, I do not have a particular exit strategy but you may want to try one that exits positions when the RSI reaches a certain level. Another one can be to exit the position using the ATR indicator to set targets and stops.
the idea to make x bars ago with a value y above RSI for Long is interesting and produce good results (optimized) when i use an exit strategy calculated too on RSI. See my backtest picture below on NQ daily.
I changed mainly the period of rsi from 14 to 2 for all RSI used. The rest is from optimization. However, becarefull, the overall sets presents overoptimized.
Hello Sofien,
But why 55 bars ago and level 50 ?
thx
Hi Jean,
The 50% level is the neutrality level. Think of it as a zero line, anything above it is a mostly bullish regime and anything below it is a bearish regime. It's the mid-point of the indicator.
The 55-bars ago comes from a technique I use to improve objectivity of the lookback periods I use. Basically, we can use, any number between 2 and infinity, which makes it tough to justify why we chose a certain number of periods. I use Fibonacci sequence to delegate this subjectivity. Hence, I may use 8, 13, 21, 34, 55, 89, etc... to at least not have to choose between a larger list which may include (10, 15, 20, 25, 30, 40, 60, 70, 80, 100, 200, 300, etc...).
Hope this helps.
Yes i see,
thank you for your answer Sofien.
Too, do you have any suggestion to exit about the signal ? in the term to complete the cycle, and to create some backtests of this idea ?
Exit strategies are numerous and they're the ones that determine a good strat from a bad one. On this one, I do not have a particular exit strategy but you may want to try one that exits positions when the RSI reaches a certain level. Another one can be to exit the position using the ATR indicator to set targets and stops.
Hi Sofien,
the idea to make x bars ago with a value y above RSI for Long is interesting and produce good results (optimized) when i use an exit strategy calculated too on RSI. See my backtest picture below on NQ daily.
https://gyazo.com/873252ecb9e3976c97e26347959c0307
I changed mainly the period of rsi from 14 to 2 for all RSI used. The rest is from optimization. However, becarefull, the overall sets presents overoptimized.
Looks good! Thanks for sharing
Yes.
No problem :-)