Re:
R & D in flow method some ideas
Friends, this thread is rocking like good old days of Saints 60 min flow--- and most of our old friends, stalwarts and new ones are now keeping tabs on this one. I have a few ideas, which can be explored, if found interesting enough.
I am a great believer in Fibonacci numbers and it is a well known fact that prices move in Fibonacci ratios. The latest range of nifty is between 61.8% and 23.6% of the high made on Jan 6, 2009 and low made on Jan 23, 2009.
Now we are playing 60 minutes time frame, which is giving good results. Now a mass of traders playing 60 minutes pivots will have more or less identical set of stops and hence trade entries which is bound to be noticed sooner or later by the market players. We have already seen this before and as this method catches on, this problem may become acute.
Now the area to do research is to find other time frames and my favourite numbers are fibonacci numbers so we can explore the applicability of this method on TF like 34, 55, 89, 144 minutes and even 233 minutes (as 4 hours TF is quite popular in Forex trading). Higher TFs are more suitable for commodity trading and will be less stressful.
The points to be checked while back testing various TFs are:
1. Do pivots behave well?
2. Are spikes more or less in a particular time frame?
3. Do filters work well i.e., whipsaws are less?
4. As prices move in waves of Fibonacci ratios, will Fibonacci number time frame behave in a better way?
5. Also if any other time frame (non-Fibonacci number) for any particular security behaves better?
This will throw up limitless possibilities and each trader will have his own favourite time frame to work on and so the mass of identical orders and hence the predictable behaviour of a set of traders will then be dispersed.
Saint, your comments are welcome. If any such work has already been done then what was the outcome. Is this worth following up?
Thanks
Bee