www.meerutmoneymana
From: Vasanth Mohan G Buddaan <bgvmohan60@gmail.
To: Technical-Investor@
Sent: Sun, 15 November, 2009 8:25:01 AM
Subject: Re: [Technical-Investor
Jayakrishnan & MMM,
A simple moving average cross over is bound to work extremely well in markets which are trending - as demonstrated by the stats of trades of Mr Sriram provided earlier by him. Obviously they are bound to whipsaw in trading (ranged) markets. Faster the system - meaning smaller the tick (bar) size and smaller the time frame, more will be the whipsaws. Ofcourse the smaller time frame Moving Average Cross Over system will pick the trend much earlier.
The next question would be same as MMM asked - what to choose ? Obviously we are interested to choose the one that gives the best results for which one may test the system for past assuming that future is going to be similar to past - which is a major assumption. Supposing one ends up with a long period of ranged market right in the beginning then the high number of whipsaws could be a mental agony and one might even discontinue the system. One obviously needs to rigid / adament and have matching staying power to take some losses (even continuous ones) like Mr Sriram.
If we can not live with the best looking lady in town due to temperamental issues then what is the point in marrying her - I don't know how it is to you but at 50 running I can understand it well enough. So taking the temperament decides what cross over one goes for : If one hates "frequent" losses - like I do - actually I hate all sorts of losses :-) then a slower system is quite ok and if one wants to be quick on the trigger and does not mind the number of losses till they are small then a faster system is the one for him. The major disadvantage of a slower system is that the entries would be delayed - at times quite delayed that one could become skeptical (the first time I traded by a mechanical system the sell came after a fall of 90+ points in nifty which was 100 + something when high of the day is considered) and so will also be the exits meaning that they will be far from respective extremes. More importanly, the draw down could also be larger - if one is not having a stop loss meaning a simple exit but adopting a 'stop & reversal' type of (continuous position) trading.
An improvement to this would be to introduce another 'indicator' which controls entries & exits with a view to eliminate the whipsawing period - the original objective being to capture TRENDS. And the system referred by you attempts that by looking at Stochastis though you have not defined how exactly the entry would be based on stochastics or for that matter the exit.
However by definition the later condition becomes the critical one as it controls the entries & exits and the entry / exit LAG would be that of the "second condition of the system" - here stochastics. In a way it might be made to make sense as these 2 indicators work best in 2 different types of markets - one (MACO) in trending market and another (stocastics) in a trading market. If the purpose of trading is to capture trending phases as against scalping / contrarian trades of a ranged market (which alone can be the objective here because of a simple MACO system) then stochastics is to be used to filter out trading ranges. Hence the entry / exit rules of stochastics have a greater bearing. You might very well know that to be able to use stochastics for contraian trades then one needs the strength of a trend and not the information of 'presence of a trend'. MACO simply tries to suggest presence of a trend either 'up' or 'down' all the time and not the strength of the trend. Lower trend values could mean a trading market where stochastics works best.
Now to go for a larger time frame for this would be like adding a larger lag to the entry / exit for the essentially a MACO system. You have also suggested that exits should only be on a 30 min time frame - presumably (?) on stochastics. One might be able to reduce number of whipsaws (but not reduce size of draw downs) but I can not understand now one can 'exit' early.
We might be able to discuss it better if you can explain the system more clearly.
gv
Dear Sir,
I came across a day trading strategy in the net. The set up was with MA cross overs and stochastics.
The author suggests to enter in 5 minute chart ema cross over signal and stochastic confirmation in both 5 minute and 30 minute charts. If stoch doesn't give bullish signal in 30 minutes chart he asks to wait.
So multi time frame seems to do a fairly good job. For entry he uses lesser time frame (5 min). And for exit the longer time frame (30 min). By exiting in the long time frame chart, one can eliminate whipsaws to great extent and avoid early exit, I believe.
Regards,
Jayakrishnan.
Yahoo! India has a new look. Take a sneak peek.
http://finance.groups.yahoo.com/group/Technical-Investor/files/!Forum Rules of Conduct.txt
No comments:
Post a Comment