Hidden divergence - page 55

New comment
 
Bookkeeper писал (а) >>

It's a good thing I didn't get away in time.

You seem to have some real work to do? If it's not too much of a mystery, soap me up a doggie, please. My soapbox is on my profile. Only if it doesn't scare you - if you like it, I'll disfigure it a bit. Right now I'm bored with "normal" drawing, so I'm just taking it easy. The drawing on the right is the candles as is, the one on the left is after my mockery. And I'm using them to draw the indices. By the way - I am only joking on forums, I am quite a decent person in correspondence.

I wanted to show that strange as it may seem - the signals are the same, only it's easier to recognize them.

And in general - just a question for coders: does anyone have some fun stuff, which calculations are based on the proportions of shadows and the carcass? Share them, if you're not sorry.

What is this mockery of candles (very interesting)? If you want to share it with us, it's better to have it printed on the left side...

 
Korey писал (а) >>

to ANG3110

The first hurdle is to correctly subtract the trend from the price, and that's where your development is very strong,

- Is it not possible to normalize periods of indicators?

That they take the trend out of the price - that's for sure. I tried to use both averages and regression but I failed to obtain the values they suggest. And in general I have a suspicion that they calculate not the correlation coefficient but the determination coefficient R^2. It would be cool to be able to calculate the correlation and divergence as they go along and set the optimal period without any adjustment using the tester. But optimal periods vary so much... I tried to optimize periods in the tester from 1 day to a week with one day step. But the results were very unattractive. And of course in that market from February till about 2 weeks ago the divergence had the best results, but unfortunately it is a tester adjustment. In reality the frequency of market fluctuations is changing and there is no way to catch this viper.

 

to ANG3110

Thank you! very valuable experience - I'll save in this direction!
-my attempts to optimize on the fly only worsen the result (although the attempts themselves were not enough)))
it seems that heuristic methods, i.e. programmatic approaches - are simply useless, and behind the curtain some theories are used.

 
Korey писал (а) >>

to ANG3110

Thank you! Very valuable experience - I will save on this one!
-my attempts to optimize on the fly only worsen the results (the truth is that the attempts themselves were not enough))
it seems that heuristic methods, i.e. programmatic approaches - are simply useless, and behind the curtain some theories are used.

This is behind the curtain: the Americans made a fuss about Georgia and now they probably use dollars for big purchases, probably weapons. As soon as this fuss started in Georgia, the dollar went up, the growth speed during the last week is really high, now the divergence is not even necessary. The cruder methods like crossing the averages and similar trend lines are working fine. But how long will these abnormal phenomena last?

 
three months. evil clowns don't give up that easily.
 
Helen писал (а) >>

Drawdowns? If without control of the deal (pending in advance, absence from the workplace), then up to 180, in my memory, is on the pound. Doubling of the deposit in a month is stable. What is your situation?

180 in what units? You guys are millionaires. Double in a month? Stable? All right, guys, I'm leaving in shame.

The figure of 3 years of background checks was tentative.

 
ANG3110 писал (а) >>

Behind the curtain - the Americans made a fuss about Georgia and there are probably large purchases now, probably weapons for dollars, as soon as this fuss started in Georgia, the dollar soared, the rate of growth during the last week is jet, - now the divergence is not even needed. The cruder methods like crossing the averages and similar trend lines are working fine. But how long will these anomalous phenomena last?

China has reduced its oil consumption by 7%. Big changes are coming.

Looks like there was a conspiracy - Russia was provoked and in a very primitive way. Like a computer game and a book by an American who foresaw 9/11 and the events in the Caucasus in them exactly August 2008. And the president there is Dmitry Arbatov ...

 
OZ0 писал (а) >> The figure of 3 years of background checks was conditional.

So 3 months count that too conditionally ))))

 
this is the best prediction)))
 
Bookkeeper писал (а) >>

And in general, just a question for coders: does anyone have some fun ideas that calculate the proportions of shadows and the carcass? Share them, if you don't mind.

{
      T1 = MathAbs(Close[i+1] - Open[i+1]);//ТЕЛО СВЕЧИ
      D1 = (High[i+1] - Low[i+1]);//ДИАПАЗОН СВЕЧИ
      R1 = T1/D1;//ОТНОШЕНИЕ ТЕЛА СВЕЧИ (текущей или n-1 зависит от i) К ЕЕ ДИАПАЗОНУ 
 
      T2 = MathAbs(Close[i+2] - Open[i+2]);//ТЕЛО СВЕЧИ n-1
      D2 = (High[i+2] - Low[i+2]);//ДИАПАЗОН СВЕЧИ n-1
      T3 = MathAbs(Close[i+3] - Open[i+3]);//ТЕЛО СВЕЧИ n-2
      D3 = (High[i+3] - Low[i+3]);//ДИАПАЗОН СВЕЧИ n-2
      T2.3 = T2 + T3;//СУММА ТЕЛ 2-ОЙ И 3-ЕЙ ПРЕДШЕСТВУЮЩИХ СВЕЧЕЙ 
      D2.3 = D2 + D3;//СУММА ДИАПАЗОНОВ 2-ОЙ И 3-ЕЙ ПРЕДШЕСТВУЮЩИХ СВЕЧЕЙ
      R2.3 = T2.3/D2.3;//ОТНОШЕНИЕ СУММ ТЕЛ 2-ОЙ И 3-ЕЙ ПРЕДШЕСТВУЮЩИХ СВЕЧЕЙ К СУММЕ ИХ ДИАПАЗОНОВ
 
      //K.OZ = (R2.3 - R1);//, 
      //K.OZ = (R1/R2.3);//,  
      K.OZ = 1 - MathAbs(R1 - R2.3);//РАЗНИЦА МЕЖДУ ОТНОШЕНИЕМ ТЕЛА СВЕЧИ (текущей или n-1 зависит от i) 
                                    //К ЕЕ ДИАПАЗОНУ, И, ОТНОШЕНИЕМ СУММ ТЕЛ 2-ОЙ И 3-ЕЙ ПРЕДШЕСТВУЮЩИХ СВЕЧЕЙ 
                                    //К СУММЕ ИХ ДИАПАЗОНОВ - ЧЕМ МЕНЬШЕ (Т.К. ОТ 1 ОТНЯЛ ВЫРАЖЕНИЕ), ТЕМ 
                                    //ВЕРОЯТНЕЕ КОРРЕКЦИЯ (РАЗВОРОТ)
)))                   
      K.OZ1 = (((T1-(T2+T3))-(D1-(D2+D3)))/(D1-(D2+D3)));//*1000*Point;//КОЭФФИЦИЕНТ ИМПУЛЬСА
      
//КОЭФФИЦИЕНТ СЖАТИЯ СВЕЧЕЙ (ВНУТРЕННИЕ МОДЕЛИ ТРЕУГОЛЬНИКОВ ИЛИ ЗАТУХАЮЩИХ КОЛЕБАНИЙ)      
      CO = (Close[i+1] - Open[i+1]);   
      HL = (High[i+1] - Low[i+1]);
      K.V =  (CO/HL);// * Volume[i+1]*Point;//    Коэфф.сжатия ТЕКУЩЕЙ СВЕЧИ - КССn
      CO2 = (Close[i+2] - Open[i+2]);   
      HL2 = (High[i+2] - Low[i+2]);
      K.V2 =  (CO2/HL2);// * Volume[i+1]*Point;   Коэфф.сжатия ПРЕДЫДУЩЕЙ СВЕЧИ   - КССn-1
      }
1...484950515253545556575859606162...66
New comment
Reason: