If I underst you Ken then this is another way to write the equation:
[ (High-Open) + (Close-Low) ] / [2 * (High-Low)]
I hacked an indictor I had inserted that equation I think it works correctly this is what it looks like I have manually written the values on the daily bars with the values I worked out the equation they line up with the indicator. See what you think.
Dean.
Edit number 2 : if the High Low were the same value the indicator would try divide by zero stop working I added a nicer fix than the last fix by checking before it tried to divide by zero I uploaded the fixed indicator here. Use at your own risk as I'm very new to programing.
if (High[i]-Low[i]==0) MomBuffer[i]=0.50; // check before trying to divide by zero
else MomBuffer[i]=((High[i]-Open[i]) + (Close[i]-Low[i])) / (2*(High[i]-Low[i]));
Dean You've got the formula exactly right I can see the results on the daily charts look promising. This works on bars/cles of almost any duration but I've found that it only provides the set-up the lower the time frame the more the whipsaws are produced. Also on daily intra-day time frames finding clear-cut entry signals seems to be a tricky proposition.
This is why I lengthened the timeframe to weekly. It provided more time to analyse the move more pips when successful the pivots of daily highs lows provided more obvious entry points with fewer whipsaws. The problem is that as a swing trading model based on a weekly bar there just isn't enough action.
I think the solution might be in a compromise between the daily the weekly time frames. I've been looking at high/low/open/close data spanning 3 days I've found I get a lot more signals instead of just doing the monday morning thing. I've combined the OBOSI idea with the Smash Day concept put forward by Larry Williams. More specifically The Naked Smash Day Reversal. A Smash Day is a day that makes a new 5-8 day high/low. A Naked Smash is when the day closes above the previous high. These are unreliable as continuation patterns but very reliable for reversals. The idea is that you sell the low of the previous naked Bullish Naked Smash Day or buy the high of the previous Bearish Naked Smash Day. It does cost in profits the length of the smash day itself but it also eliminates the great majority of the false starts. It also allows me to stick with one of my most beloved trading principles. NEVER TRY TO PICK THE VERY BEGINNING OR THE VERY END OF ANY MOVE.Jesse Livermore said it better "One of the most helpful things that anybody can learn is to give up trying to catch the last eighth - or the first. These are most expensive eighths in the world"
To give you a live example of the a Naked Smash Day Reversal on friday 10 March USD/CHF was overbought. That friday also was a Bullish Naked Smash Day so I placed a sell order at its low at 1.3123.
The order was executed on the Tuesday 14 by Friday 17 March it closed at 1.2900.
209 pips profit after the spread. Last week It worked again going the opposite way but this time I manually trailed a stop got stopped out at 50 pips profit. The risk management model needs work. I've been using a 50 pip stop but perhaps I should be using the the far extreme of the smash day as the poitn at which I abon the trade. I've been using 50 pips because its all my bank will allow for without further reducing my investment or risking more than 5% of the bank on any given trade.
I need to test the 3 days model I need to test other ps. All my work so far has been been on the Swiss. I lack the coding skills to do mass longterm backtesting. If you are able to assist or know anyone who could that would be fantastic.
According to my calculation the Swiss is overbought again so the entry point should be for a short trade should be the low of the previous bullish smash day - thursday 23/3/06 - 1.3047. Please let me know if there's any part of the theory that I haven't explained clearly.
I know the risk trade management need work but I think combining the OBOSI with Naked Smash Day Reversals has a lot of potential for building a simple robust trading method
Gotta go.
A system based on price action is not oscillators or moving averages. Breakouts or breakdowns out of a trading range is one way to trade price. There are many more. It just takes many years of learning the hard way. Ouch!! Price tops every other way of looking at the market.
Dean Unfortunately I don't know how to attach a chart. Last weeks entry point was 1.2946 this week's is 1.3084. This is the high of the previous smash down. 1.2946 last week was the low of the previous smash up. Place a 50 pip stop a 50 pip profit take on your limit order. Then wait.
actually i have been researching on systems based on price action for quite some time.
i have found out that SUPPORT & RESISTANCE AVERAGE TRUE RANGE (ATR) are 2 tools that provide a lot of assistance in building such systems.
i believe that a good system should have 3 characteristics to sustain in the the long run:
1. positive probability
2. robustness
3. risk management
overall i have found ATR to be the best tool to be incorporated in a system to make it adaptable to changing market dynamics (robustness).
you can change your position sizing for a trade based on the ATR during that period. this allows one to manage risk in the most efficient manner.
however i have not been able to find enough material on how to build SUPP. & RESI. levels.
Let me explain my earlier point. Trading based on price action is just that. It is not systems or indicators. It's pure simple following what the market is doing. Support resistance in whatever time frame one is trading as well as longer term sup/res levels which is just based on you got it price action. No calculations needed. Now if anyone is discussing RSI or MA's or Wave counts etc. that is not price action. There are derivatives. Too late to change this thread name but please don't refer to it as price action. OK.
This might not be what you are after but have you seen SHI_Channels it's a dynamic Support Resistance Channel used on MetaTrader only thing is you can not back test it if you are that way inclined as it is constantly changing. Check it out over at: http://fxovereasy.50webs.com/Indicators.html
Licensed Brokers
Re: Building a system based on price action
Linear Mode
