RSI based trading strategies

Quote from frostengine:

Anyone have ideas on other ways RSI may be traded? The next two ideas I will be researching are:

#1 Using RSI to find divergence
#2 Finding patterns within RSI movement.. using several AI techniques

Any more basic ideas to try before taking it to that level?

First of all most people do not know that the RSI value depends on how many data points they have in the price history:

http://en.wikipedia.org/wiki/Relative_Strength_Index#Cutler.27s_RSI

Sorry if I spoiled the party.
 
I just finished testing for bullish hidden divergence on the RSI. This was a tough one to test for because finding divergence programatically is a bit challenging. However, the results turned out pretty well.

The method produced the best results I have seen so for in the RSI only testing phase. I can see several ways this may be useful in a strategy down the road. On my blog I posted the full source code as well as a few charts showing the performance. They can be found here: http://thestrategictrader.com/?p=274

Next I will be looking at regular divergence to see if there is an edge there as well.

I notice a lot of people are negative on RSI in general. That is the purpose of this adventure. I want to remove the emotional connection from these indicators and just test them. See what works and what does'nt. So far I am finding several valid edges. Its not perfect, but there does appear to be some uses for it.
 
Quote from intradaybill:

Did you read the reviews?

Yes I did. She does give pointers to the Composite indicator in the book, but does not give it out because she dislikes the intellectually lazy and will not support the trader welfare state. I could not agree more.

The formula is there. She gives the ingredients to the composite and enuf charts to reverse engineer the thing. And it is a great oscillator especially if you want to trade with the RSI.

The RSI section has valuable info if you want to learn to use RSI to determine trend and strength of trend.
 
Quote from frostengine:

I just finished testing for bullish hidden divergence on the RSI.

...So far I am finding several valid edges. Its not perfect, but there does appear to be some uses for it.

I'm curious...what stock did you test this indicator on that you said the statistics show several valid edges ?

Edit - I just found the below earlier quote by you that answers my question.

Mark

Quote from frostengine:

...After mapping out which values are best, I took one of the better combinations RSI(3) < 30 and tested it against the 30 DOW stocks to see how robust the method was. Keep in mind the values were optimized against ES. The method was able to pass our target percent profitable thresholds in 17 of the 30 DOW stocks and pass the target profit in 7. This leads me to believe this method is fairly robust...
 
RSI is a "range oscillator"... as are Stochastics, CCI, Williams %R. They all work in a similar way... and the "range" need not be horizontal.

Spend some time with them so that you get to know how they correlate to price in various markets. They all give false signals in the "wrong" kind of market, so you should know what those are before trading them. Vary the number of periods to perhaps find better short-term correlation to price.

The "range oscillators" have value if you know how they usually correlate and how they fail..... none of them work all of the time.
 
Back
Top