From theory to practice - page 1835

 
Maxim Kuznetsov:

For the "correlation matrix" - you have to do a topologistical-order (if I am not mistaken in the English transcription), and obviously based on fundamental data such as turnover and balances.

They change slowly, but clearly affect the limits of rate fluctuations. Without such a thing, the whole matrix is just a picture

You may try to look for cointegration between exchange rates and trade balances. For example, take the EURUSD exchange rate and the balance sheets of the USA and the Eurozone. You may use the API of the local calendar to calculate the balances. We will have to calculate them in R. We have to find a way to reduce balance sheets and exchange rates to a "common denominator" - it won't work in a "raw" form.

I do not see a special practical sense in this, but for an article (about the calendar and cointegration) may be suitable.

 
Aleksey Nikolayev:

You could try to look for a cointegration of exchange rates and trade balances. For example, take the EURUSD exchange rate and the US and Eurozone balances. The balances can be taken from the API of the local calendar. We will have to calculate them in R. We have to find a way to reduce balance sheets and exchange rates to a "common denominator" - it won't work in a "raw" form.

I do not see a special practical sense in this, but for the article (about the calendar and cointegration) may be suitable.

They are not and cannot be cointegrated - most of the Forex market turnover is speculative and does not serve real foreign trade operations.

And not to "reduce to a common denominator", but to standardise.

Shame on you

 
Aleksey Nikolayev:

You could try to look for a cointegration of exchange rates and trade balances. For example, take the EURUSD exchange rate and the US and Eurozone balances. The balances can be taken from the API of the local calendar. We will have to calculate them in R. We have to find a way to reduce balance sheets and exchange rates to a "common denominator" - it won't work in a "raw" form.

I don't see a special practical sense in it, but it may work for an article (about calendar and cointegration).

What does R have to do with it... it has nothing to do with anything.

you have to tie the technical analysis (which is 99% ready to work on the average hospital temperature) to the foundation.

As a sensible thought - rates move as-yet-unknown, but their amplitudes have limits - trade balances and their balances. Because, after all, it is not stocks, but currencies from which everything else is measured

 

Some song excerpts


Thank you, friend!

 
Дмитрий:

Shame on you

Your signature is correct.

 
Maxim Kuznetsov:

What has R got to do with it... it has nothing to do with it at all.

You have to link technical analysis (which is 99% ready to work on average hospital temperature as well) to the fundamentals.

As a reasonable thought - rates move as-yet-unknown-how, but their amplitudes have limits - trade balances and their balances. Because it is not stocks, but currencies from which everything else is measured.

I don't see any contradiction between R and technical analysis. R is needed so as not to get bogged down at the stage of reinventing bicycles.

You can try to take the amplitude, the volatility, the trend or something else - the main thing is that the series obtained from the quotes were "proportionate" to the balance series (which are known at a limited number of points in time and change more slowly). This is what I call a common denominator.

 
khorosh:
I am always surprised by critics of martingale when they take the most primitive option without loss limits as an example and prove that martingale is bad.
It's not that it's bad, it just doesn't make sense.
 
khorosh:
I am always surprised by critics of martingale when they take the most primitive option without loss limits as an example and prove that martingale is bad. But there are a lot of variants of TS that use lot increase. Those who have been practicing it for a long time know that there are at least dozens of them. The critics of martingale must first study all these possible variants and work with their testing and then draw conclusions.
I'm surprised anyone else is still rooting for martin
 
Aleksey Nikolayev:

I don't see any contradiction between R and theanalysis. R is needed to avoid getting bogged down at the stage of reinventing bicycles.

You can try to take amplitude as well as volatility, trend or anything else - the main thing is that the series obtained from quotes are "commensurate" with the balance series (which are known at a limited number of points in time and change more slowly). This is what I call a common denominator.

There are a couple of moments left to dissect another naïveté.

 
Uladzimir Izerski:

There are a couple of moments left before another naive girl is dismembered.

Careful with the starch, Volodya.