Good (Sunday) morning everyone!!!
As some of you may know I posted on Friday with regard to the CSI problem that I thought I had solved but after close inspection it turns out that I was talking utter nonsense so I deleted the post!!!
However: I NOW think that I have INDEED solved ‘half the problem’ with the CSI as it relates to (most) forex pairs.
At the moment I am ignoring all the other variables in the CSI equation i.e. I’m only using the ADXR(14) and ATR(14) values simply to find the forex pairs that have good strong directional movement as well as high volatility (for reasons mentioned in ‘the book’). I’m calling this my ‘CSIADXADXR(14)’ (as opposed to ‘CSI’ purely to distinguish between my ‘watered down version’ of the CSI and the ‘proper’ version). In other words: the ‘CSIADXADXR(14)’ is giving you the forex pairs that have good strong directional movement as well as high volatility but not necessarily the forex pairs to be traded that are the ‘best bang for buck’ as it were.
I’ve also come to the conclusion that you CANNOT mix forex pairs, commodities, and equities together by using this method for the simple reason that with forex pairs the margin requirements, spreads, commissions payable, and $ per pip movement are all very similar to each other BUT with commodities and equities this is NOT so i.e. the leverage is totally different (much lower), the margin requirements and commisisons payable are a lot higher, and the $ per pip movement is totally different not only between forex pairs but between the different commodities and equities themselves.
In short: I have arrived at the conclusion that it’s OK to eliminate the margin requirements, spreads, commissions payable, and $ per pip movement when rating forex pairs against each other thus the ‘CSIADXADXR(14)’ formula BUT if you’re going to rate forex pairs against commodities and equities as well then you HAVE to use the FULL CSI equation.
I have attached an updated (based on Fridays close) ‘Daily ADX ADXR CSIADXADXR’ work sheet for you to look at. The instruments that are ‘greyed out’ are the instruments for which there is still some further investigation necessary before I’m happy with their overall rating i.e. I think that the ENTIRE CSI equation has to be applied to them before an accurate overall rating against the ‘normal’ forex pairs can be arrived at (but I’ll get to that tomorrow).
Take a look.
If I’m right (which I’m almost 100% sure that I am) then it simplifies things somewhat i.e. all the trend following systems are to be used with instruments / pairs with the HIGHEST CSIADXADXR(14) values and the systems to be used with range bound instruments / pairs are to be used with the instruments / pairs with the LOWEST ADX(14) and / or ADXR(14) values.
I’m also happy that the JPY crosses are now being correctly rated against the other pairs. The method was simple i.e. ATR(14) is giving you the Average True Range IN POINTS / PIPS so all you have to do is agree the ATR(14) between the different pairs to get an accurate reading and this is done by multiplying the ATR(14) by a ‘multiplier’ e.g. for the JPY crosses (where prices are quoted with two decimal places) the ‘multiplier’ is 100 whereas with all of the other forex pairs (where the prices are quoted with four decimal places) the ‘multiplier’ is 10 000. This is then used for the CSIADXADXR(14) or the CSI calculation.
(By the way: for those of you who saw this message earlier I’ve just edited it i.e. you DO NOT divide the resultant ATR(14) by 100 after multiplying by the ‘multiplier’ of 10 000 i.e. if you do it ‘disallows’ you from comparing other intruments like the stock index futures).
Daily ADX ADXR CSIADXR.pdf (12.4 KB)