Market theory - page 175

 
Roman Shiredchenko:

Yeah I can see that... :-)

He is, as usual - STRONGLY on his wavelength (TF D1, take/stop size is close as if there were none at all)...

Let him ride that wave. At least there are no Martins. And there is a match. Respect.
 
Yousufkhodja Sultonov:

No, I'm still waiting for the results. 1 daily bar is formed in 1-1.5 hours on the computer. Maybe I'm doing it wrong, but I'm waiting for the results.

Alexei, nothing can make serious adjustments to the model anymore, as you put it. I noticed that you also hesitated for a moment in the right direction of your research, but you turned down the path of your opponents in time, which is good to see.

I'm all for that. The main thing is to catch a pattern.
 
There's something to that) the spikes do warn of a trend change.
 
Artyom Trishkin:

No, in order to test by minute openings, we first need to add to the Expert Advisor the operation by the given timeframe, not by the one on which it is installed.

And, I am not sure that the indicator will give the correct results based on minute openings - it uses close prices of bars (20 by default), and the current price at the current chart - also, by the way, Close

Yes, the indicator should work at closing prices of that timeframe, to which it is applied, and the closing condition is tested on the fact of tick arrival (of a minute or whatever). This should be considered at design stage for accurate testing.
 
Алексей:
Yes, the indicator should work at closing prices of the timeframe it is applied to, and the closing condition should be tested on the fact of tick arrival (minutes or whatever). This should be considered at design stage for accurate testing purposes.
If we test by the open prices, then how will the indicator receive the ticks inside the bar (even a minute one), if there are only four prices in the bar - OHLC ?
 
JohnPawn:
Dear Yusufkhoja, Just to clarify, compare two charts, one based on your data and the other based on MA(20) for 2010. Intersection with the current price at the same point in time.
Intersections and should coincide by definition. For example, what is the Leo level, is a geometric mean value of price levels Tsr (Leo) = (Ts*P)^0.5, and what is the Leopard level is the arithmetic mean value of these levels Tsr (Leopard) = (Ts+P)/2. At the moments of trend changes Leo gathers all prices at the same level and therefore Ts1(Bears)=Csr(Leopard)=Csr(MA)=P=C2(Bulls)=Copt(Leo). Therefore, the coincidence with the MA level once again confirms the validity of the present market theory. And in between, these levels behave quite differently as they are prescribed by their regularities, according to the many times given ratios, with computer accuracy. Even in the moments of surges and a significant deviation of levels from their average values within the ticks, all levels strictly observe the specified subordination.
 
Artyom Trishkin:
If it is tested at open prices, how will the indicator receive ticks inside the bar (even one-minute), if there are only four prices in the bar - OHLC?
I have not seen his code, but I think that it is not necessary to look inside one-minute bars. I have made an EA on 1-minute bars and tested it by open prices only, and I have set the processing of market entry once an hour, for example.

And its logic should be that once a day a new indicator point is formed (since it looks at 20 days). But in fact it is recalculated on each tick. In short, nothing is clear. Yusuf should clarify the logic of his TS.
 

So far I got this result on M5, EUR/USD, with 0.01 lot and TP = SL = 45pp (five digits) from start of the month 03. 07. 15 to now 13-20 MSK 20. 07. 15:

Bars in test 3438
Ticks modelled 6761
Modelling quality n/a
Mismatched charts errors 0
Initial deposit 1000.00
Spread Current (2)
Total net profit 3837.89
Gross profit 7489.38
Gross loss -3651.49
Profit factor 2.05
Expected payoff 1.40
Absolute drawdown 548.35
Maximal drawdown 1516.56 (41.01%)
Relative drawdown 69.40% (1096.90)
Total trades 2734
Short positions (won %) 2077 (63.17%)
Long positions (won %) 657 (74.73%)
Profit trades (% of total) 1803 (65.95%)
Loss trades (% of total) 931 (34.05%)
Largest
profit trade 4.50
loss trade -4.62
Average
profit trade 4.15
loss trade -3.92
Maximum
consecutive wins (profit in money) 299 (1339.74)
consecutive losses (loss in money) 229 (-1031.32)
Maximal
consecutive profit (count of wins) 1339.74 (299)
consecutive loss (count of losses) -1031.32 (229)
Average
consecutive wins 55
consecutive losses 29

 

Congratulations. Nice work.

Yusuf, please explain the idea. What are you going to get as a result?

Since we are all empathetic, you already have results and great ones, but what does that mean?

For example. We study the daily bar and get the real price of the previous day ???

Or, increase the calculation period to 20 years and get the real price one year ahead ???

What do you get as a result. You can see the indicator data in the photo, but the date and time of the forecast is not there!!!! I don't need the codes, my processor is too slow.

If it's possible, it will be in the market price format, the data will be more accurate.

For example EURUSD Date price. Is this possible?

My respect,

Yuriy.

 
Yura3512:

Congratulations. Nice work.

Yusuf, please explain the idea. What are you going to get as a result?

Since we are all empathetic, you already have results and great ones, but what does that mean?

For example. We study the daily bar and get the real price of the previous day ???

Or, increase the calculation period to 20 years and get the real price for the year ahead ???

What do you get as a result. you can see the indicator data in the photo, but the date and time of the forecast is not there!!!! I don't need it, I don't have a processor that can handle it.

Sincerely,

Yuri.

I want to make sure that the tester results are confirmed in a real account.