Monday, June 05, 2006

Free Nifty Trading system which outperforms buy and hold

This is one of the most rudimentary systems I have ever developed. I started with moving average systems last year. I currently use 7/105 on 10 mins.
The yahoo version is free!!!

So
Why 7/105? Because those are the values I got after running an optimization
Also usable are combinations in the region of 5-8/78-121.
I also use systems based on gann swing/ganntrend/ewt/multiple timeframe fibonacci for risk management (targets/stoplosses). If you want live nifty targets just ask me on yahoo. I am usually online during market hours and i'll usually have a target/stoploss in mind for nifty :). (yahooid=rajak1981)

Gist of the system.
1)If 70 min avg is above 1050 min avg play long, if 70 is below 1050 play short.
In pictoral terms: if green is above orange then long; if orange is above green then short.
2)try to use pullbacks to the green line the 70 min ema or the black line 210 ema to enter positions. (210 ema will result in smallest stoplosses under most circumstances, just look at the charts in an uptrend 210=great support and in a downtrend 210 = great resistance. 70ema can sometimes get whipsawed.)
3)Do not short if rsi is under 30
4)The stoploss is 0.8% to be employed as soon as you enter your trade.






Here is a summary
Statistics
All trades Long trades Short trades

Initial capital 500000.00 500000.00 500000.00

Ending capital 5157103.43 3750550.76 1906552.67

Net Profit 4657103.43 3250550.76 1406552.67

Net Profit % 931.42 % 650.11 % 281.31 %

Exposure % 96.90 % 64.87 % 32.03 %

Net Risk Adjusted Return % 961.19 % 1002.17 % 878.20 %

Annual Return % 938.05 % 654.27 % 282.72 %

Risk Adjusted Return % 968.03 % 1008.59 % 882.58 %

This is the performance over the last year.
Basically 938% returns on margin capital (20% margin)

Note: Also linking to full backtests from 2001-2006
with leverage 2001-2006 (annual return 663%,2845801.25 % net)


without leverage 2001-2006 (annual return 53.21%, net return 1223%)



The non leverage returns are most significant because they show that this silly system can return 53.21% per year without leverage under current market conditions
Ps: Doing buy and hold on nifty over the same duration would have resulted in a net return of only 240% or so. 1000-3400 = 240% roughly.

(I expect this system to stay profitable as long as we stay in this same cycle degree impulse. Which by my counts should be for ATLEAST another 2-3 years. Maybe another 10-20, depending on what your wave count is w3 of 3 into all degrees? or w5 somewhere :))


A lot of technicians these days post systems without specifying the profitability of the system. I think this is in bad taste. Someone posted a system using fib averages. The trailing stop employed by the system made sure that the only person getting rich would be the broker and not the trader. Another person posted a basic 12/26 macd system , 12/26 macd is not even profitable in 3,4,5 min timeframes. Even on 10 min timeframes it barely gives 446% per year.
Before I post a system I will make sure it is profitable and reasonably optimized. However I always suggest that people do their own testing before using a system. This will instill some confidence and allow the trader to use the system properly.
In most cases it is not the system that fails or stops working in a market. It is usually the trader who stops following the system after a series of losses. This usually happens right before a massively profitable trade is caught by the system.
Always remember that the most exciting moves will usually come out of boring sideways/consolidating markets.


Note:
If you want a detailed tutorial+afl on how to use this system/how to enter/use trailing stops to minimize entry/exit signals then it'll cost you 2500 Rs.