Alternative and common approaches in the construction of TC - page 3

New comment
 
hrenfx:

........

Example (on a martin):

Common:

You analyse some specific financial instruments as possible uses for a martin. You select the best of them and run your martin.

- You use history to select the parameters for a particular financial instrument. You execute your martin and successfully drain it.

Alternative:

First, you must clearly define how the financial instrument must behave for a martin to work - 1.

- You take an arbitrary variation of Martin parameters.

Then you need to think how to create such a synthetic tool from the current financial instruments that would meet the conditions (1).

- From the current financial instruments create a synthetic that shows the profitability of the margin.

Then you figure it out and execute your martin on the synthetic.

- You start your martin on the synthetic and successfully drain it.

Comparison:

At this example it is obvious that the alternative will give much better results - stability and profit.

- The alternative will give exactly the same result - a stable loss, but how many prestige it has.

 
Mischek:

Maybe you can give me another example, I didn't understand the Martin example.


There are many martins, so no details.

 
Azerus:

You have apparently written without thinking deeply.

Any TS will (in theory) fail! Whether it's Martin or something else is irrelevant.

As long as you know the distributions of the non return movements of the financial instrument over 20 years. And it's elementary to find it out by analyzing the history. Then, taking into account this distribution, you can create a martin, that won't fail during these 20 years of history.

You can say that it will fail anyway. And you will be right. Just as you will be right about not only the martin, but any TS fitted to 20 years of history.

But unlike many TS, a Martin is a stable system (sloping straight line) with little variation (if adequately adjusted). And the probability that a martin will shed is equal to the probability that the FI will change its behaviour significantly compared to its behaviour over the last 20 years.

What does substantially mean? It means that the distribution of no-return movements will become different.

So, for the same martin, you can create a synthetic with the best distribution of no-roll movements. And trade on that synthetic.

 
 

to Mischek: no need to be weak, the discussion continues.....

to hrenfx:

- if we choose parameters for a TS by the history of one instrument, it does not mean that these parameters will be optimal for this instrument in the future....

- if we choose some synthetic by arbitrary parameters, the history of which will show good results for the given parameters of the TS, it does not mean that these parameters will turn out to be optimal for this synthetic in the future....

- there is no difference between the probability/probability of changing the behaviour of an instrument and the behaviour of the synthetic in the future (i.e. the synthetic can change its behaviour in the same way as a simple instrument) .....

 
Azerus:

to Mischek: no need to be weak, the discussion continues.....

to hrenfx:

- if we use the history of one instrument to select parameters for TS, it does not mean that these parameters will be optimal for this instrument in the future....

- if we choose some synthetic by arbitrary parameters, the history of which will show good results for the given parameters of the TS, it does not mean that these parameters will turn out to be optimal for this synthetic in the future....

- there is no difference between the probability/probability of changing the behaviour of an instrument and the behaviour of the synthetic in the future (i.e. the synthetic can change its behaviour in the same way as a simple instrument) .....


OK, then explain what the parameters are if the TS works without indicators
 
Azerus:

- if we use the history of one instrument to select parameters for a TS, it does not mean that these parameters will be optimal for this instrument in the future....

- if we use arbitrary parameters to select some synthetic, the history of which will show good results for the given parameters of the TS, it does not mean that these parameters will turn out to be optimal for this synthetic in the future....

- there is no difference between the probability/probability of changing the behaviour of an instrument and the behaviour of the synthetic in the future (i.e. the synthetic can change its behaviour in the same way as a simple instrument) .....

Quite right.
 
hrenfx:
Absolutely right.

Then why did you start this thread?
 
Azerus:

Then why did you start this thread?

To gossip ), to post their dogmas, to argue with everyone about how wrong they are, etc.

SZS: Of the last 3 threads, one makes any sense at all.

Oh yeah, I also forgot: to engage the intellectual part of the forum users

 

In short. Author, have you heard about the multicurrency? )))

In fact, you propose the following scheme: EA --- function of creating a synthetic trading tool from the market mn-o-meter --- market. In terms of form, it is a multi-currency EA.

The essence of the idea (as I understand it) is to separate the logical synthesis block of a trading tool. In my opinion, it is a sensible idea. Respect.

12345678910
New comment