Correct calculation of currency indices. - page 24

New comment
 
excelf:
Brilliant, no where have I looked for something like this, thank you!

Here's an implemented version.
 
2 trollolo: Vadim is who?
 
trollolo:


It is not the absolute value of the index that matters (it will not do anything in static conditions - the relationship is rigid), but the change in its movement (character).


And I will refer you to my indicator. There are both absolute and relative movements.
 
Mathemat:
Vadim is who?
it's a hint of a personal relationship. if the new corish troll222 goes back to his old ways, he'll go the same way.
 
trollolo: Well, I'll go to Kramorov's branch, because they don't like newbies here and react to the truth with harassment(

What harassment... Calm down. Go on, talk interestingly.
 
trollolo:

It surprises me that some people put the right tool for measuring in the first place, ......

I have been dealing with the issue of calculation of currency indexes for a long time, it seems to me that certain results can be achieved, but in this case we will still have a lag (even with the use of neuristics) rate formation is so messy that it seems to me not quite right to separate a synthetic index from pairs of allies. The presence of orders such as DND, the use of currency not only as paper, but also to pay for certain contracts (i.e. rather to monitor the commodity markets is needed ) . Of course it is possible to collect all available instruments (to find certain weight coefficients from them) all of it ))))), in general it is deeply imho not necessary to complicate an already difficult task. And if we are going to approach the question systematically, create a set of procedures processing current market fragment (in our case currency, we will have lag anyway), we will seem that we lack computer power, we will start to write code for it and it will cause lag anyway))). Because the point is where the data comes from. If I scan only half a sheet of text and then try to decode it, that's where .... Oh my god !!! there will only be half of the text. Summary. Imho, it is much more efficient to create a euro and dollar index - using only one currency eur/usd
 
solar:
Imho, it is much more efficient to create a euro and dollar index - using only one currency eur/usd


Bummer!

Is it OK that in this case there is nothing to create these very indices from?

 
Zhunko, the style is very different, take a closer look.
 

trollolo: вот кого имел ввиду.

Something completely amorphous, inarticulate and uninteresting, I couldn't stand it and didn't finish watching it. Only what does it have to do with this forum and with you?

 
trollolo:


1. This is logical. But it is reasonable for, as you say, pure indices. the point is probably that the index can be different - the basis for its movement is different. if you calculate as you do - it is no better and no worse than Semenych. there is statically one and the same thing. And if you initially use not the absolute prices but their characteristics separately - reduced to a common form, you should get something else - something like a graph not of price changes but changes in the nature of movement.

2. Forget, why do you think that it is a pure index (as you put it), does anyone have an original, in relation to which there is a comparison?

3. pure index cannot exist in principle, it can be the most complete basket of currencies.

1. I have a precise terminology about it. All index calculation methods that can't be called clean, but which can be very useful for trading, are called unclean. I am not against them, I just say that before inventing and using them, one should learn how to calculate pure ones. And I assert that not all forex programmers know how to do it. That is all. Semen Semenych's epic is a vivid example. He publicly developed a line of cluster indicators and at the same time he publicly confirmed the non-existence of the correct oscillation point for the quotation of all the basket currencies relatively to it. The only compromise he saw as not entirely correct (which I completely agree with) was to fix (equate to one, for example) one of the currencies. (Of course, he rejected this incomplete possibility. And rightly so.) And no one from the programmer people corrected him, although there were many people on the Alpari forum where all this was taking place, and especially in his thread. But the reference point does exist - it is a geometric mean of all currencies, which cannot be separated from the basket and calculated as a separate value (so it was not discovered?), but which, despite its virtuality, can be a realistic measure for all currencies in the basket. And a 100% correct one at that.

2. Yes, it has an original, a model and a benchmark. It is the simple property of identical equality to one of the product of all the indices of the basket at each point in time, plus, of course, also the identical equality of the mutual ratio of any pair of indices to the quotation of the corresponding currency pair (again, at each point in time). Find another way to calculate "indices" possessing this constant property and I will immediately remove the title "pure" from my method. :) Go ahead, I'm very interested to read it.

3. oh yes, size does matter. And yet for any size, the concepts of clean/unclean remain relevant. Such is the hygiene fact. :)

1...171819202122232425
New comment
Reason: