backtesting vs real trades

syswizard said:
Good idea. IMHO He would need to test with 10 different symbols to get the testing time down to one year.
More...

I feel better now:) just answered above, I do use same strategy, with the same parameters, on multiple assets except I use 15min, 30min, 60min charts
 
I just checked, I have about 695 trades with 30 symbols and tested for 18months
but the fact is, avg net profit in a given 6months is so diff from -PnL to +PnL as I gave in Original post. so is this a bad strategy to go LIVE and work on tweaking? to get +PnL
 
DTorSwing said:
1st 6 months $3351
2nd 6 months -$4250
3rd 6months $1452

I havent checked each month. I am usually interested in yearly PnL for any strategy, as I get an average of 12months
More...

How many profitable/unprofitable trades? What are the $ values of your average winner/loser?
 
here is whole summary of 30 symbols, run in Ninja's strategy optimiser
2 choices

1) choice 1 - 45min chart
Total # of Trades : 345 (commission paid $694)
% profitable 45.53 %
158 winning trades over 189
avg trade $31.17, total PnL $10,814

2) choice 2 - 12min chart
Total # of Trades :1238 (commission paid $2476)
% profitable 39.82 %
493 winning trades over 745
avg trade $6.83, total PnL $8454

anyone will say, I love the choice1 , due to its super avg per trade $31/
and due to low # of trades of 45min chart, i can trade more symbols as I am left with more margin !! smart idea ha :)

what do you all think
 
IAS_LLC said:
According to simple sampling theory (emphasis on simple... maybe too simple), for a strategy to be statistically significant with a confidence level of 99% and a margin of error of 5%... You would need to conduct at least 666 trades, and 363 of them would have to be profitable to reject the null hypothesis.

If you wanted to reduce the margin or error to 1%, than you would need 16641 trades, with 8470 profitable to reject the null hypothesis.
More...

In the other thread you used 30 trades. State your formula. You know no statistics. This is not coin tossing.
 
IAS_LLC said:
Well consider this:

Lets say you can test 100 trades, and want to do 99% confidence. This would mean your margin of error is about 13%. To reject the null hypothesis, you would need to have 61/100 profitable trades. But there is a stochastic element to the market, and the backtest is only accurate to 61 +/- 13 trades, 1 Sigma. A one standard deviation error is not an unlikely event... It is reasonably likely that your 61% win rate is actually a 48% win rate.

With only 100 trades, you really can't reject the null hypothesis reliably. Even with 666 trades, it is difficult.

If you want to trade systematically, I would recommend shelling out a bit of cash for decent historical database. Otherwise, your backtests will not be statistically significant, when margin of error is factored in.
More...

Write down you formula for those claims. Confirmatory statistics do not apply to backtests. You seem not to understand this. Your statistics knowledge is too limited. Only real trades can be used for those tests. By the time you get 600 trades you may be already broke. Do you understand this or you will continue with absurdities?

I repeat for you. Statistical testing for hypothetical results makes no sense.
 
Sergio77 said:
Write down you formula for those claims. Confirmatory statistics do not apply to backtests. You seem not to understand this. Your statistics knowledge is too limited. Only real trades can be used for those tests. By the time you get 600 trades you may be already broke. Do you understand this or you will continue with absurdities?

I repeat for you. Statistical testing for hypothetical results makes no sense.
More...

sorry, were you asking me or the other person who responded
 
IAS_LLC said:
According to simple sampling theory (emphasis on simple... maybe too simple), for a strategy to be statistically significant with a confidence level of 99% and a margin of error of 5%... You would need to conduct at least 666 trades, and 363 of them would have to be profitable to reject the null hypothesis.

If you wanted to reduce the margin or error to 1%, than you would need 16641 trades, with 8470 profitable to reject the null hypothesis.
More...

How did you come up with these numbers?
 
You must log in or register to reply here.
Back
Top