Hi guys, I came back to you and am making my way from page 127 to the last one.
Hopwfully I can get updated and then post more replies.
See you around.
Hopwfully I can get updated and then post more replies.
See you around.
10 Ema and 200 Ema cross system 6 replies
EMA crossover (ema 14 & 34) 4 replies
Please help code an EMA Crossover 12ema x 24 ema cross 1 reply
EMA's and MA's confiming small time correctoin ema's crossing 1 reply
DislikedI leave it to your thoughts and comments.
Oh, and I almost forgot… here’s the Indicator with the changes, and also the MTF Stochastic it uses (it has to be in your custom indicator's folder).Ignored
Quoting ronmatzkinDislikedBefore i even test the indicator, before I read the comments of the other members and before I look into the code - well done and thank you.
I know you made a hugh research and learned a lot - coding and logically wise.
Well done and thanks again!Ignored
Quoting toniraiDislikedI think I might have a possible solution/alternative to the Stochastic H4... I thought of it yesterday night, but I have to make a little change to the code and run a test, to see if it will do what we want.
At the same time, I plan to test the "no stop & reverse" thing.
Hopefully I will do it tonight.. if not, tomorrow for sure.Ignored
QuoteDislikedCalculation
The Stochastic Oscillator has three variables:
- %K periods (Pk). This is the number of time periods used in %K calculation. By default is 5; (if in 4H timeframe we use 5, in the 1H we'll use 20)
- %K Slowing Periods (Sk). This value controls the internal smoothing of %K. A value of 1 is considered a fast stochastic; a value of 3 is considered a slow stochastic. By default is 3;
- %D periods (Pd). This is the number of time periods used when calculating a moving average of %K. By default is 3; (if in 4H timeframe we use 3, in the 1H we'll use 12)
The formula for %K is:
%K = 100*SUM (CLOSE - MIN (LOW, Pk), Sk) / SUM (MAX (HIGH, Pk) - MIN (LOW, Pk)), Sk)
Where:
CLOSE — is today’s closing price; (same value in both configurations)
MIN (LOW, Pk) — is the lowest low in Pk periods; (as we keep the same distance from present candle with Pk=20, it's the same value)
MAX (HIGH, Pk) — is the highest high in Pk periods; (as we keep the same distance from present candle with Pk=20, it's the same value)
SUM (CLOSE - MIN (LOW, Pk), Sk) — amount composed CLOSE - MIN (LOW, Pk) for period Sk;
SUM (MAX (HIGH, Pk) - MIN (LOW, Pk)), Sk) — amount composed HIGH (Pk)) - MIN (LOW, Pk) for period Sk.
The only value that makes the calculation non-proportional in time is Sk, but by trying, I find 6 fits the purpose of giving close crossings as in the original 4H timeframe
QuoteDislikedimho we're going in the direction that we can't do better than macd>/<0 with stoch h4.
double sto_main = iCustom(NULL, 0, "#MTF_Stochastic", Stoch_Period, Stoch_KPeriod, Stoch_DPeriod, Stoch_Slowing, Stoch_MA_Method, Stoch_PriceField, 0, i + 1); double sto_signal = iCustom(NULL, 0, "#MTF_Stochastic", Stoch_Period, Stoch_KPeriod, Stoch_DPeriod, Stoch_Slowing, Stoch_MA_Method, Stoch_PriceField, 1, i + 1);
DislikedL = 257 (O_O...)
TP4 = 29
TP3 = 47
TP2 = 55
TP1 = 151
(72.5+94+68.75+37.75-257)*4*23 = 1472
... we probably have to think something else... I don't understand why results are so different with this version... everything is the same apart from stochastic that should be very very similar.
imho we're going in the direction that we can't do better than macd>/<0 with stoch h4.
we should probably leave it demo testing 2 months to see if the repaints are a big problem or if it is marginal... meanwhile working on the stoch h1 to find the best "proxy" to the h4 systemIgnored
DislikedI've noticed huge difference in performance between Indication and written EA.
The system was scoring great because we've known the value 3 hours earlier !!! I think it's dangerous to do performance testing with indicators.
Looks like we have to stick to 1H stochastic.Ignored
DislikedAnother case: the Indicator is not including spread in its calculations. Eg. for EURUSD we have about 3000 pips profit (from 2008.01.15), but there where more than 1600 transactions. If we have 2 pips spread, then we've lost money (3000 - 1600*2 = -200) !!!Ignored
Dislikedwith your settings I had 6000+ pips about.
I never tried to switch ATR multiplier I always thuoght the best is 1.0 as we always used till now.
edit: with 2.2 I got 8000+ results
ok now this system even avoids repaints...Ignored
Dislikedwe really need to have the same broker here -__-
I still use oanda but soon I'll switch to a metatrader one. Can you me tell me yours?Ignored