World currencies index (clearly visible as the bubble burst) - page 3

 
IgorM:

What do you use for analysis: ticks, OHLC or something else?

Imho it is hard to find correlation between different instruments as market orders are not correlated with each other in any way, I believe that there is no multi-currency dependence on ticks


I took closing prices
 
Kolivi:

I took closing prices.


OK, already some constructiveness in the topic, I should note at once that Close is no better/worse than Open, because the closing of a bar is accompanied by the opening of a new one, as well as the TF is an input discreteness of time intervals

Have you taken into account the weighting of each pair against one currency? After all, a 1 pip change in EURUSD is much more "expensive" than USDCHF for example

 

Nikolai, no one is discouraging you from your trading method. Just introduce weighting coefficients for each pair. Above I gave you a link to a thread where the percentages between pairs in the total world money flow are given. Insert them into your formula .You get a new more correct curve. The machine will calculate everything on its own by closing or by opening. The line will be more reliable, but the trading method will be the same. Check the difference between these curves for the sake of interest.

 

Think about it, you have 28% of the world's EUR/USD turnover and only 14% of the EUR/Yen. But the euro is worth 1.3 quid and the quid is worth 82 yen. All pairs without yen can now be taken out of the formula. Their share in the total is not more than 1-2%. And all the rest are crosses with JPY, which is not true, if you are talking about the pairs (currencies) potfal and can be justified by trading crosses with JPY only, but it is questionable. Do not be lazy to check it, because you only need to enter into the formula the coefficients, and the machine will calculate them. Or you can rub our noses in it, or you will understand your wrongness.

 
gss:

Nikolai, no one is discouraging you from your trading method. Just introduce weighting coefficients for each pair. Above I gave you a link to a thread where the percentages between pairs in the total world money flow are given. Insert them into your formula .You get a new more correct curve. The machine will work the same way, whether it closes or opens. The line will be more reliable, but the trading method will be the same. Check the difference between these curves for the sake of interest.

This approach is good for looking at the dynamics of economies, but it is not good for trading. The weights have to be dynamic. I wrote about this above.
 
gss:

Think about it, you have 28% of the world's EUR/USD turnover and only 14% of the EUR/Yen. But the euro is worth 1.3 quid and the quid is worth 82 yen. All pairs without yen can now be taken out of the formula. Their share in the total is not more than 1-2%. And all the rest are crosses with JPY, which is not true, if you are talking about the pairs (currencies) potfal and can be justified by trading crosses with JPY only, but it is questionable. Do not be lazy to check it, because you only need to enter into the formula the coefficients, and the machine will calculate them. Or you will rub our noses in it, or you will understand why you're wrong.


I don't understand you.

You want me to give 28% of portfolio to EUR/USD, 14% to USD/JPY, etc.?

Or should I screw the volume to the price index? That is, if A=EURUSD, B=USDJPY, Z-another pair, then (A*0.28+B*0.14+....Z*0.01)/28=Index?

In the 2nd case I see no sense, it will simply change the value of the resulting index, but the chart will remain the same.

In 1 case the sense of trading all 28 instruments is lost, it is easier to trade with 1 pair, as other pairs will not be able to compensate for losses of eur/usd

Or did you have something else in mind?

 
Here is an Excel spreadsheet
Files:
 

Schizokos on the forum.

It will be interesting to see how it is you will buy all 8 currencies at the same time. Even if with different weights. Against ZAR ? Or NOK?

Oh yes! You're going to buy pairs . 28 pieces. Well, well. Don't forget to cut back on the fractions, if you're smart enough.

Good luck, of course.

 

1. trading a pOrtfolio is not easier than 1 pair, but much harder. One mistake and your pOrtfle's bottom will fall off.

2. If you don't pick pairs, and you pile everything into a purse, it's not a purse, it's a bin or a hobo's bag.

3. Dealing with 8 indices is easier than dealing with 28 currency pairs.

4. Multiplying pairs by some ratios is a mistake. The trading volumes of a currency pair or the share of a currency in the total weight of all the world's paper has nothing to do with the exchange rate.

5. Good indices must be INDEPENDENT, I would even say ORTOHONAL, otherwise they will do more harm than good.

6. All pairs are not needed to calculate indexes, the number of indexes-1 is enough. Arbitrage is a different subject and there are no fish there.

 
AlexeyFX:

What are you implying? I'm talking about O.

Reason: