"Miracle", "digital" "group" movement indicator - page 4

 
trol222:
And how does the formula take into account the changing weights when calculating the currency indices?

Respect! :-)) First post on the forum and quite important and on topic! I can boast that this issue has been resolved for me.

It's imperative to take into account changes in weighting factors. Otherwise these currency indexes are useless. You won't get a prediction of the pair's behaviour.

The formula is not simple. I do not want to distribute it. When you figure it out for yourself, don't post it.

 
trol222:
And how does the formula take into account changing weights when calculating the currency indices?

It's simple, open and implemented. See Recycle.
 
Zhunko:

Looked it up... :-))) Thank you! Was smiling while reading it. You didn't get it either. It turns out it's simpler for me. Significantly simpler.

Certainly didn't guess how the person saying he can count thinks...

:-))) A common mistake, everyone is trying to find weighting coefficients associated with some kind of attachment to the economy. These coefficients reflect the real situation in the economy, but are not suitable for trading.
Where do you see a link to the economy?! Have you read between the lines? A universal mathematical method for finding BP relationships and weighting factors in these relationships has been described and implemented. What does the economy have to do with anything? This method is suitable for any BPs (even the BP equitability of various TSs).
 
Zhunko:

Looked it up... :-))) Thank you! Was smiling while reading it. You didn't get it either. It turns out it's simpler for me. Much simpler.

:-))) The common mistake, everybody is trying to find weighting coefficients connected with some links to economy. These coefficients reflect the real situation in the economy, but are not suitable for trading.

You could just tell me.

Only a few would really understand.

From these few, most likely, no one will be able to benefit, as is almost always the case.

I, for example, really like the Sart indicator itself - ta....- simple and clear.

 

Zhunko

And how many currency pairs do you need at least to calculate the weights of these pairs, and should you limit your calculations to currency pairs or use other instruments from the list of available ones? By the way, maybe you can achieve the same success when applying it to other instruments - not currency pairs (indices, commodities, stocks, etc.) and also trade them successfully?

 
Zhunko:

I use all pairs available in the terminal. I am constantly looking for brokers and brokerage companies with the largest number of currency pairs. The combination of currency pairs in the index also gives interesting results.

Right now Exness is leading in the number of instruments. They have 168 instruments.

Not enough majors for you? All crosses are linearly connected to the majors. So they don't contain any additional information. Why the hell use crosses?

Support your words calculating the weights with something other than "I know, but I won't tell anyone".

 

Zhunko

How much load will be put on the computer (the terminal) in this calculation if we don't take into account the synchronisation of ticks of different currencies (let's use minutes) and "omit" the problem of holes?

 
Zhunko:

I don't understand about the crosses... The calculation of currency indices uses pairs containing that currency. What does this have to do with crosses? If I need MXN, I will use pairs that contain MXN. I do not care what is the name of the group of this currency pair.

MXN is MXN. So we have data on these major currencies:

EURUSD, GBPUSD, AUDUSD, NZDUSD, USDCHF, USDCAD, USDJPY, USDMXN, ....

There are no crosses at all. It is claimed that this data is sufficient to get any pairs with MXN currency:

EURMXN, GBPMXN, AUDMXN, NZDMXN, MXNCHF, MXNCAD. MXNJPY, USDMXN, ....

There is no doubt in this statement. So once again, there is no need to analyse "168" currency pairs, just analyse the majors.

 
hrenfx:

MXN, so MXN. So, we have data on the following majors:

EURUSD, GBPUSD, AUDUSD, NZDUSD, USDCHF, USDCAD, USDJPY, USDMXN, ....

There are no crosses at all. It is claimed that this data is sufficient to get any pairs with MXN currency:

EURMXN, GBPMXN, AUDMXN, NZDMXN, MXNCHF, MXNCAD. MXNJPY, USDMXN, ....

There is no doubt in this statement. So once again, it is not necessary to analyse "168" currency pairs, it is sufficient to analyse the majors.

The formulas for calculating other pairs with majors I know. If there is a tight interdependence, nothing will go anywhere.

I personally try to have a complete currency matrix as well.

 

it seems to me that we are coming to https://www.mql5.com/ru/forum/125734 funny how history moves.

Then there is the question of whether it is easier to find the dependencies of currency pairs on each other or to calculate true exchange rates from a basket of pairs and then use the true exchange rates to predict the probability of a pair's movement.

In the first case you need weighting coefficients for assembling the basket.

In the second case, there is a reference pair from which currency rates are calculated.

Reason: