Random wandering - page 3

 
Let's generate a series of stock quotes, using any available random number generator.

This may be the clue here. Not every generated series is really random, many generators fall into cycles or vice versa - into trends, therefore it is really possible to show profit on the vulnerabilities of such series, but what is the point of it? So, the only way is to play with it. No one (well, like a casino) will give such a free series to trade on it, there are no fools - a normal stable generator will be attached. And on the market it (the row) itself is not free.

 
I gave a link to an article in the beginning. If you can't make money on sb, then refute it.
 
I'm not saying anything about the market, I'm only interested in the sb
 
ILYA_365:
I gave a link to an article in the beginning. If you can't make money on sb, then refute it.
You can make money. You can't make money.
 
ILYA_365:
I gave a link to the article in the beginning. If you can't make money on SB, then refute it.

You can't. The trick of such experiments is to build another SB in the form of "equity" of the TS going upwards as if it were possible. But in reality, the average increase in such an equity will be negligible (tens of times less than the real costs in the form of spread and commission). Therefore, a fine equal to the commission of any SB, because again, the average increment of any, even the most gleefully going up SB, is negligible.

 

As I understood from the article, the stop is twice as long as the take. As a result we have a lot of small profitable trades and few big losses. If the sample is finite, it is possible to achieve that in a particular run the lots are rare and there is an excess of profit. However, if the sample is aimed at infinity, i.e. to make profit always, everything will come to naught, because the total profit and total loss are equally probable.


That is, over a long distance, the sum of all losses will be just equal to the sum of all profits, with any stop and take variations. Everything is balanced in nature.

 
Aleksei Stepanenko:

As I understood from the article, the stop is twice as long as the take. As a result we have a lot of small profitable trades and few big losses. If the sample is finite, it is possible to achieve that in a particular run the lots are rare and there is an excess of profit. However, if the sample is aimed at infinity, i.e. to make profit always, everything will come to zero, because the total profit and total loss are equally probable.


That is, the sum of all losses will be just equal to the sum of all profits, with all stop and take variations. Everything is balanced in nature.

There's a picture of a test of 30 random "securities" there. Everywhere in the +

 
You know, you can choose successful runs and show them by keeping the truth to yourself.
 
Vasiliy Sokolov:

You can't. The trick of such experiments is to build another SB in the form of "equity" of the TS going upwards as if it were possible. But in reality, the average increase in such an equity will be negligible (tens of times less than the real costs in the form of spread and commission). Therefore, the penalty on a trade equal to the commission of any SB, because again, the average increase of any, even the most eagerly going up SB is negligible.

Do I understand you correctly that it is possible to make money on SB, but the costs will eventually lead to minus?

 
Aleksei Stepanenko:
You know, you can choose successful runs and show them by keeping the truth to yourself.

You can, but it's not convincing