For those who have (are) seriously engaged in co-movement analysis of financial instruments (> 2) - page 22

New comment
 
CC is not even remotely substantiated. It was simply a gimmick in its day as one of the first multi-currency indicators.
 
hrenfx:
CC is not even marginally justified. It was just a gimmick at one time as one of the first multi-currency indicators.

Let's talk about why you don't think it's justified. I have arguments - or rather questions:

1. Smoothing with mashups - is it necessary? Why iron out something that should be more or less stable for some reasonably long time?

2. Why play to the downside and not to the upside? Even a currency with a max index can continue to attack for a while longer, and for quite some time.

3. Why is such a strange function from pair movements chosen - the sum of the signs? There's certainly no statobased reasoning here...

Besides, I can easily prove that if not smoothing, then at any given moment there will be a Leader currency (rising relative to all others in the full set of pairs) and a Loser currency (falling). Accordingly, the best entry would be to buy Leader/Loser (sell Loser/Leader). This is not good, because there is such an entry at any time. But pairs are not in the momentum all the time, they happen to be in flat... We need a reasonable level of minimal movement which can be considered as such, and everything below it should be cut off as a zero movement.

 
Mathemat:

Let's talk about why you think it is not justified.

To me SS is seen as a self-defeating shamanism with charts. From its description:

It should be noted that originally the Complex_Pair indicator was developed, but it was quite resource-intensive. Later during the research it was discovered another principle of making the same indicator, but using very fast algorithm that allows extracting the same information, but only from the price chart of separate currency. Despite the fact that Complex_Pair and Complex_Pair1 obtain the data in a completely different way, the visual similarity between them was absolute on higher timeframes. On smaller timeframes, there were some differences, but Complex_Pair1, can work with any financial instrument, be it stocks, futures or commodities.

How can we even talk about multicurrency here! When multicurrency MAs are the difference between two monocurrencies.

If we are not talking about the implementation of SS itself, but about the idea of finding the "strongest" and the "weakest" is another matter.

Or to rephrase: ranking currencies by strength.

 

So, summary.

1. SS is shamanism. I remember this remark by Semenych . And it suggests that there doesn't seem to be any multicurrency there. Perhaps the whole point of the problem is in the calculation algorithm?

2. Synthetics in the stationarity (channeling) calculation - no justification. Empiricism.

3. what else can you think of? Somehow, I think there was a post by IgorM. But it was an offtopic, as the post concerned the arbitrability on a single FI. Suppose we take two indicators the difference of which is more or less returned to the average and open when this difference is too high. I haven't tested it.

 
In spite of the fact that Complex_Pair and Complex_Pair1 receive data absolutely differently, the visual similarity of them on higher timeframes was absolute. On smaller timeframes, there were differences, but Complex_Pair1, can work with any financial instrument, be it stocks, futures or commodities.

it seems he (CC) made a mistake in the indicator code, there was a break command missing, in the switch calculations for TF....

===

switch(Period())
     {
       case 1:     res += iMA(sym, tf, per*k, ma_shift, Mode, Price, i); k += 5;
       case 5:     res += iMA(sym, tf, per*k, ma_shift, Mode, Price, i); k += 3;
       case 15:    res += iMA(sym, tf, per*k, ma_shift, Mode, Price, i); k += 2;
       case 30:    res += iMA(sym, tf, per*k, ma_shift, Mode, Price, i); k += 2;
       case 60:    res += iMA(sym, tf, per*k, ma_shift, Mode, Price, i); k += 4;
       case 240:   res += iMA(sym, tf, per*k, ma_shift, Mode, Price, i); k += 6;
       case 1440:  res += iMA(sym, tf, per*k, ma_shift, Mode, Price, i); k += 4;
       case 10080: res += iMA(sym, tf, per*k, ma_shift, Mode, Price, i); k += 4;
       case 43200: res += iMA(sym, tf, per*k, ma_shift, Mode, Price, i);
     }
   return(res);
 
Mathemat:

3. what else can you think of? There was a post by IgorM once, I think. But it was an off-topic, because the post concerned the arbitrability of a single FI. Like we take two indices, the difference of which more or less returns to the average, and open when this difference is off the scale. I haven't tested it.

Offtopic:

The MA is a great indicator. Below is the EURUSD and exponential average difference and its distribution:

Great to trade this... if it were not for one BUT: the MA cannot be traded.

That is, you cannot buy EURUSD and sell its MA at the same time. Otherwise it would be a gravy pair trading.

 

You can't do it all at once, of course. You can accumulate a swing for a while and then, at a profitable moment, take it and lock it with a reverse price trade. I've been watching the stats. Sometimes it's not so good (it can take a long time to accumulate, i.e. wait for the right moment). There are some theorems in statarbitrage theory about the timing of rate reversals under general assumptions about the nature of the process.

But that's offtopic.

P.S. I have a couple of ideas about how to try to cross Semenych with synthetics. I'll try to formalise some later.

 
Mathemat:

3. what else can you think of? There was a post by IgorM once, I think. But it was an off-topic, because the post was about arbitrability on a single FI. Like we take two indices, the difference of which more or less returns to the average, and open when this difference is off the scale. I haven't tested it.


an old theme - I don't use it anymore, it's about "candlestick catching" - candlesticks are caught, but when FI "calms down" this strategy starts to lose deposits during mini-trend changes

i now analyze the FI as a market mechanism - i see much more dependencies, both related to bar closing time and to the best price for the period

for sabbath:imho - a profitable strategy should not work out by the mechanism proposed by topikstarterom, the only thing you can take from stat.Arbitrage is to make a prediction from which major dollar flowed into the other - ie find a currency that will depend ONLY on the rate of their crosses, but not on the dollar, I think that's particularly look for nothing - usually the yen serves as a buffer in the excess of dollar supply in European currencies

 

For those who have (are) seriously engaged in analysis of the co-movement of fintechs (> 2).

A strange search for strange synthetics.

These "synthetics" are already plentiful amongst the staff instruments.

You can always find "synthetics" among them, which hold normal channels on the diaries for months at a time.

 
More, let's have something concrete in the studio. Let's have a look together.
1...151617181920212223242526272829...38
New comment