Is the Cauchy difference a precursor to a reversal and/or correction?
Dear programmers. I propose to test this hypothesis. As I stated earlier https://www.mql5.com/ru/forum/58256, the arithmetic mean (MA) and geometric mean (MG) price difference, which I called "Cauchy difference"-"K" in the real market of goods and services directly affects the profit received. It is also shown that price movements are organised around two break-even points defined by MA and MG, i.e. real (current) and virtual (market) prices.
Assumption or hypothesis: The Cauchy differential K must reverse in advance before reversing from one trading mode around the first break-even point (zone) to the second break-even point (zone).
If you help me quickly put together a program to test this hypothesis, I will give simple formulas for programming and testing this assumption right here. I did on Exel and there seems to be something to it.
You can see that, Cauchy's Difference reverses before the price. Subsequently confirmed the reversal. Again, so far everything is at the level of speculation. When the indicator is available, it will be clear.
Yusuf, good day! What about the antelope? Did the tiger get mauled?
what kind of indicator is needed ?
what kind of indicator is needed ?
The indicator that draws the second figure is the Cauchy difference with the possibility of changing the period N, like the MA indicator. Last column of the table. Now I think how to adapt this indicator for placing it on the price chart, for now - place it in the basement of the price chart. If there is anything that is not clear in the mechanism of calculation, ask - I will answer. Thank you for your prompt reaction.
first you need a formula to write the indicator
this is from your posts quote -
Louis Cauchy proved that the arithmetic mean of two quantities (Sar.) is always greater than their geometric mean (Sgeom.), where Sar. = (x1+x2)/2 and Sgeom. = (x1*x2)^0.5. The difference of these quantities I named "Cauchy difference" (K) after the great French mathematician, i.e. K = Sar. - Sgeom.
x is what value ?
and if done for K elements it will be a very large number when multiplied ?
or divide not by 2 but by K ???
First, you need a formula for writing the indicator
this is a quote from your posts
Louis Cauchy proved that the arithmetic mean of two quantities (Sar.) is always greater than their geometric mean (Sgeom.), where Sar. = (x1+x2)/2 and Sgeom. = (x1*x2)^0.5. The difference of these quantities I named "Cauchy difference" (K) after the great French mathematician, i.e., K = Sar. - Sgeom.
x is what quantity ?
and if you do it for K elements the multiplication will be a very large number ?
or divide by K instead of by 2 ???
Look at the sequence of the calculation scheme on the table. We do not divide the sum of prices by K, but by N - the number of data in the calculation of MA and take the root of the Nth degree from the product of prices in the calculation of MG.
Indicator formula: K = MA - MG
Look at the sequence of the calculation scheme on the table. Divide the sum of prices not by K, but by N - the amount of data when calculating the MA and take the root of the Nth degree from the product of the prices when calculating the MG.
Does the diagram match the table?
Look at the sequence of the calculation scheme on the table. Divide not by K, but by N - the amount of data when calculating MA and take the root of the Nth degree when calculating MG.
OK, then according to your table
MA = (C1+C2+...+Cn)/n
MG = (C1*C2*...*Cn)/n
K = MA - MG
In the last post I asked about the value of x, here it is C - ?
ok, then according to your table
MA = (C1+C2+...+Cn)/n
MG = (C1*C2*...*Cn)/n
K = MA - MG
In the last post I asked about the value of x, here it is C - ?
MG = (C1*C2*...*Cn)/n - not correct, correct: MG = (C1*C2*...*Cn)^(1/n) - root of nth degree of the product of prices;
K = MA - MG is correct.
Yes, x=C is the current price. For a particular bar: C = (O+H+L+C)/4
n - period of the indicator.
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Dear programmers. I propose to test this hypothesis. As I stated earlier https://www.mql5.com/ru/forum/58256, the arithmetic mean (MA) and geometric mean (MG) price difference, which I called "Cauchy difference"-"K" in the real market of goods and services directly affects the profit received. It is also shown that price movements are organised around two break-even points defined by MA and MG, i.e. real (current) and virtual (market) prices.
Assumption or hypothesis: The Cauchy differential (K) must reverse in advance before reversing from one trading mode around the first break-even point (zone) to the second break-even point (zone).
If you help me quickly put together a program to test this hypothesis, I will right here provide uncomplicated formulas for programming and testing this assumption. I've done on Exel and there seems to be something to it.
You can see that, indeed, on the M1 TF the Cauchy difference (K) reverses before the price, and the deceptive price spikes before the reversal no longer affect the indicator's verdict - decline. Subsequently the reversal is confirmed. I repeat, everything is on the level of supposition so far. When we will examine the indicator, it will become clear.
Here are the indicators "Cauchy Difference" and "Derivative of Cauchy Difference" for MT4 in normal (created by Iurii Tokman) and MT5 in normal and logarithmic scales and "Derivative of Cauchy Difference" indicator in one package created by elibrarius whom I am immensely grateful to.