From theory to practice - page 555

 

О!

I'll probably wait until Automat writes an article about ACF and then I'll go off in pursuit of the healing Grail again.

 

Here's what I was thinking...

Never mind physics - let's turn to pure mathematics.

1. So we have found Laplace on increments, let it be Erlang of order 300, analogous to OPEN M5

2. The sum of these increments will give a xy-squared with k=number of terms. In the limit it is a normal distribution. Right?

3. i.e. sum of OPEN M5 returns in a sliding window, say, 1440 such values = week, will form a normal distribution with known quantile function and expectation = 0. Is it so?

4. Neuro-networkers, I'm sure, would easily turn such a process into cash.

 
 
Alexander_K:

Here's what I was thinking...

Never mind physics - let's turn to pure mathematics.

1. So we have found Laplace on increments, let it be Erlang of order 300, analogous to OPEN M5

2. The sum of these increments will give a xy-squared with k=number of terms. In the limit it is a normal distribution. Right?

3. i.e. sum of OPEN M5 returns in a sliding window, say, 1440 such values = week, will form a normal distribution with known quantile function and expectation = 0. Is it so?

4. Neural networkers, I'm sure, would easily turn such a process into cash.

I read aboutErlang, found nothing, that connects him with trading on the market, though the market at that time was already quite well known. I only realised thatErlang was connected to rural telephones)

What do you want to take from him, even from the 300th order?

P.S. About Laplace is also interesting - space and stars. Are you going to find a pattern on star cycles? Maybe it's easier to flip a coin a few times)

 
Yuriy Asaulenko:
You may easily do it in forex.) And you say it won't work.)
Let's take the rouble as the most dynamic currency. In one year, 8 rubles has gone up, that's 4 kopecks a day. A rampant trend.

You could simulate the exchange on the forex market.

What time is the eurobucks futures clearing?

 
Alexander_K:

Ahem... Let me repeat the question - so you found a pure Laplace distribution and what? Look up Laplace motion on the internet - same jumps, same junk.

A Bas, who obviously graduated without triplets, assures that going beyond any sigma does not guarantee a return to the mean.

Only exponentially decreasing ACF gives that guarantee.

Returning to the mean alone is not enough. We need a straight line return with minimum drawdown, that is, without going into minus and triggering the SL.

 
Andrei:

One return to the average is not enough. You need a straight line return with minimum drawdown, i.e. without going into minus and triggering the SL.

A_K is also a freeloader - one of them wants a straight line, the other, you see, the flow is not right. This and that are different.
Why don't you moderate your appetite and you'll be fine)).
 

From recent signals.

If traded, the trade is on the plus side.

 
Evgeniy Chumakov:

From recent signals.

If traded, the trade is on the plus side.

Well, yes, it's beautiful.))