From theory to practice - page 743

 
_o0O:

Extremely simple example: we apply MM on martin, you argue that equity will also get SB with the same characteristics as the original SB? There will be a NEW pattern that didn't exist, won't there?


Equity will really get SB (just from martin we should look in logarithmic scale), distribution can calculate A_K, he likes it :-)

 
_o0O:

An extremely simple example: apply MM on a martin, you argue that equity will also produce an SB with the same characteristics as the original SB? There will be a NEW pattern that did not exist, won't there?

Of course SB too, possibly with different characteristics. Just build equitability on a piece-by-piece basis, not on a piece-by-piece basis. In other words, we do not need to exclude what happens inside the trades, while the severe drawdowns take place there. The regularities cannot appear there if they don't exist from the very beginning. Martin simply "postpones the end" - gives you the opportunity to tolerate these drawdowns for some time and stay in the illusion that the end will not come.

 
Yuriy Asaulenko:

This is wrong. Draw MA - it's derived from SB, and you get a lot of patterns. You can build a tester's Grail on them in no time).

Yes, I didn't put it quite right. By derivative I meant building equity, i.e. some way of slicing SB (or finryad) into pieces and summing them up.

 
secret:

Of course also SB, perhaps with different characteristics. Just build equity on a per-trade basis rather than per-trade basis. In other words, do not throw out what happens inside trades, and there are severe drawdowns there. The regularities cannot appear there if they don't exist from the very beginning. Martin simply "postpones the end" - gives you the opportunity to tolerate these drawdowns for some time and stay in the illusion that the end will not come.

... but the "law of losing a player" will come anyway ))
 
multiplicator:
... But there will still be a "law of the player's flush" )))

No, it won't). If all players really play with each other on SB as in the market, after a while all the money will go to 5-10% of players, and they will tell everyone how dashingly they win on SB.

Everything will be like in real life.))

 
multiplicator:
If I fall asleep and wake up in a hundred years and people ask me what's going on in Russia now, I'll answer: drinking and stealing

If I come to this thread a year from now, I'll see that it's still looking for the same grail).
Yuriy Asaulenko:

However, the cast will change. and for different reasons.

alexander will starve to death from constant searching, and not working. the machine will get a coin in its throat from tossing, and it will choke. transcendrimer will write on the forum from his smartphone from the factory. the new one will get married and calm down. renat, after spending 10 years on the forum, will finally deduct a profitable ts. muzichenko will engage in house building, as he likes. asaulenko will earn his 20% per month and bored on the forum.
just kidding! all names are fictitious!)

 
Yuriy Asaulenko:

No, it won't). If all players really play with each other on SB as in the market, after a while all the money will go to 5-10% of players, and they will tell everyone how they dashingly win on SB.

I think this can be illustrated even more simply - just "buy and hold" SB, and it is very likely to go one way and for a long time (arcsinus law, as I recall). For a long time, but not forever. And of course, the outcome of 100 such experiences in total will be zero)

 
secret:

I think this can be illustrated even more simply - just "buy and hold" SB and it is very likely to go one way and for a long time (arcsinus law, as I recall). For a long time, but not forever. And of course, the outcome of 100 such experiments in total will be zero)

Rather, forever.( If not forever, you will regularly be able to smell porridge burning in the kitchen a year ago.))

Yes, and the outcome is very unlikely to be zero. Even at 16 quitters, if memory serves, the probability of a zero outcome is ~17%, and with more, it tends to zero.

 
Yuriy Asaulenko:

With or without a robot, it doesn't matter.

How to train is clear, but I'm not saying it's easy. But you can't find one, that's for sure).

Isn't the LCI an option?
 

For those who have nothing better to do, post CLOSE M1 data for any year for any pair in .csv format.

Preferred file name, for example: EURGBP 2017 CLOSE M1.csv

I will check them against Variance Gamma Process and post the results here.

Thank you.