Need Your EXpert Comments

#1
Hello Members,

I have created a swing trading system and backtest on all NSE stocks. Backtest criteria are (1) All quotes (2) Position trading (3) No short trades.

Please see the result and share your views.




(moneypick)
 
#3
Has brokerage and other costs been factored in

Regards
I am using zerodha as my broker therefore I have put Rs.20 per trade as brokerage charges. Yes, there is some other charges also i.e. stt, stamp duty,turnover tax and service tax also. But I do not find the way to put these charges in amibroker analysis setting.
 

guptak03

Well-Known Member
#4
please give full details ...
#the image showing is back tested on which stock
#how many years data used
# what time frame used
#etc etc;)
 

jahan

Well-Known Member
#6
Hello Members,

I have created a swing trading system and backtest on all NSE stocks. Backtest criteria are (1) All quotes (2) Position trading (3) No short trades.

Please see the result and share your views.




(moneypick)
Hello,

Thanks for asking my opinion.......

by looking ur backtest report....the system seems good...

see ....the backtesting is done or performs only on one distribution.this the limitation of backtesting.

u need a simulator to simulate all the possible distributions of ur 337 trades.

by taking ur data i have run 100 simulation's of ur 337 trades

all are given positive results...sometimes greater capital than in backtesting and sometimes little less than ur ending capital. overall the system the system is excellent..

the max consecutive losses may come more than 5,in my simulation it showed 8 straight losses in one of the 100 distributions.(which is not a big issue, for risking 1% of ur capital).

here are the some problems with excellent systems......

1)by seeing ur annual return i.e less than 9%(in backtesting) i can guess the backtesting peroid. banks can easily give u that kind of annual returns may be even more.

2).less no.of trades per year....even u can have a loosing year in bear market condition.

3).frequency of trades does matter in trading,study vantharp books on trading systems he will explain each and every aspect that need to be considered while building a system.

..........

i have also one system which gives aout 70-75% results with near or more than 1:2 ratio,but it produces only 13-20 trades per year,i use this sytem with conjunction to other systems to increase my expectancy.....

iam not a strong believer of Fundamental analysis, i believe what i see on charts,and i also know that all glitters(charts) are not gold.

so i can't help u in fundamental analysis.

regards,
 
#7
Hello,

Thanks for asking my opinion.......

by looking ur backtest report....the system seems good...

see ....the backtesting is done or performs only on one distribution.this the limitation of backtesting.

u need a simulator to simulate all the possible distributions of ur 337 trades.

by taking ur data i have run 100 simulation's of ur 337 trades

all are given positive results...sometimes greater capital than in backtesting and sometimes little less than ur ending capital. overall the system the system is excellent..

the max consecutive losses may come more than 5,in my simulation it showed 8 straight losses in one of the 100 distributions.(which is not a big issue, for risking 1% of ur capital).

here are the some problems with excellent systems......

1)by seeing ur annual return i.e less than 9%(in backtesting) i can guess the backtesting peroid. banks can easily give u that kind of annual returns may be even more.

2).less no.of trades per year....even u can have a loosing year in bear market condition.

3).frequency of trades does matter in trading,study vantharp books on trading systems he will explain each and every aspect that need to be considered while building a system.

..........

i have also one system which gives aout 70-75% results with near or more than 1:2 ratio,but it produces only 13-20 trades per year,i use this sytem with conjunction to other systems to increase my expectancy.....

iam not a strong believer of Fundamental analysis, i believe what i see on charts,and i also know that all glitters(charts) are not gold.

so i can't help u in fundamental analysis.

regards,
Thanks for your view friend. This system show me 28 buy opportunities in this year. Yes it show 9% return but it is on my profit+investing amount therefore you see the large ending capital. If we have deposited our starting capital in the bank from 1994, is our bank ending capital equal to this system ending capital..........perhaps No.

I do not understand this "simulator to simulate all the possible distributions of ur 337 trades".

How can we do this in Amibroker.

Regards,

(moneypick)
 

jahan

Well-Known Member
#8
Thanks for your view friend. This system show me 28 buy opportunities in this year. Yes it show 9% return but it is on my profit+investing amount therefore you see the large ending capital. If we have deposited our starting capital in the bank from 1994, is our bank ending capital equal to this system ending capital..........perhaps No.

I do not understand this "simulator to simulate all the possible distributions of ur 337 trades".

How can we do this in Amibroker.

Regards,

(moneypick)
Hello,

ok, metastock doesn't calculate the way amibroker does ....so its my fault..(in 9% case).

until today no real backtesting software is available to public usage.

simulator distribution means....

lets say for ur 337 trades we assign a marble....and then by putting all those 337 marbles in a bag.

....and pull a marble each time upto all marbles are complete(this is one distribution).

again we put all marbles in the bag...and pull each one upto completion......all marbles will not come, same as first time we pulled.(this is 2nd distribution)

and now u can see that every distribution is different than other,that is what iam done...with ur 337 trades

and the limitation of backtesting is it shows the results for only one distribution.

see if ur uncomfortable with my comments then, pls ignore and don't go very deep in statistics just trade ur way if u have confidence in ur system.

"sometimes what happens is somebody's know how can stop someone great idea/concept".

regards,
 
Last edited:
#9
@moneypick : Is your stock data adjusted for splits/bonus/dividends. Otherwise a excellent idea may also show bad results ,specially since your backtest time period is large.
@Jahan : which software do you use for simulation.

Regards,
 

mastermind007

Well-Known Member
#10
Hello Members,

I have created a swing trading system and backtest on all NSE stocks. Backtest criteria are (1) All quotes (2) Position trading (3) No short trades.

Please see the result and share your views.

(moneypick)
What can I comment on this? You've not shared any technical aspects of system (and I've got no problems with it). If this results are genuine, go ahead and take a plunge. Jahan has raised good points ...
 

Similar threads