acrary could you share the specific parameters for the two MA's? I am new and would appreciate the information. Thanks Ken
yes historical not ... <font color=BLUE>futu<font color=RED>RISK</FONT>al</FONT> I remind that the turtle fund sunked not so long ago ... a bit like LTCM. And since it is result in high scale it is tighly linked with the bull's market of this century, since it is a phenomena that occurs only once maybe in 50 or 100 years it's a dice play for those who comes too late in the game ... BTW about MAs one of my past thread : http://www.elitetrader.com/vb/showthread.php?s=&postid=260515&highlight=simulation#post260515 "Simulation is an important concept. By injecting the expected results into the (MA) "models" allow to control its validity and avoid taking position against the trend."
Ah, okay. We're on the same page now and I agree with you. The Turtles did a breakout on a single MA. And there are other good systems in which a MA crossover (or other trend signaling method) on a smaller timeframe is used to trigger entry in the same direction as a larger timeframe's trend. You're right, my comments were more general.
Inputs: Length(40); Vars: Top(0), Bottom(0), Middle(0); Top = LinearRegValue( H, Length, 0 ); Bottom = LinearRegValue( L, Length, 0 ); Middle = LinearRegValue( ((C+L+H)/3), Length, 0 ); If c >= Top Then Begin SellShort Next bar H Limit; End; If c <= Bottom Then Begin Buy Next Bar at L Limit; End; If MarketPosition = 1 And (C >= Middle or Close > AvgEntryPrice) Then Begin Sell This bar on Close; End; If MarketPosition = -1 And (C <= Middle or Close < AvgEntryPrice) Then Begin BuyToCover This bar on Close; End; If MarketPosition = 1 Then Begin Sell Next Bar at (AvgEntryPrice + 0.01) Limit; End; If MarketPosition = -1 Then Begin BuyToCover Next Bar at (AvgEntryPrice - 0.01) Limit; End;
Great strategy if true? At 11:30 on days the market is at or close to the Highs SELL with a stop just above the HOD. At 11:30 on days the market is at or close to the Lows BUY with a stop just below the LOD. 90% success rate right ?
I've been running this system for 8 months real money on one of our test accounts. We use 10 stock portfolio, we spread capital evenly between them, then we enter initially in each stock with only 10% of capital allowing 10 time pyramiding. We averaged around 83.4% accuracy. P.S. Be careful with stocks you choose! They have to be at least 2.5 beta!
You're overlooking one thing: the high or low is in with 90% probability. Key is to find which one will hold, the high or the low.
I know. I always present the bare bones ideas leaving the room for you guys to experiment. In real life we always use something more tuned (we use B/A analysis to refine the system). Have Fun!