A great book on testing and optimisation - page 9

 
FOXXXi писал(а) >> I think out-of-sample, forward and all that crap is clinical.

What's not a clinical case?

 
LeoV >> :

What's not a clinical case?

So are you still willing to argue with this or fully admit that it's true without any buts?

 
FOXXXi писал(а) >>

So are you still prepared to argue with this or do you fully admit that it is true without any buts and buts?

I'm not arguing, as you can see, I'm just asking -

LeoV wrote :>>What 's not a clinical case?
I'm curious - and that's fine. A normal question to your quote.
 
LeoV >> :

I agree. I have come across that there is also an 'out-of-sample' fit. And this fit is harder to understand and avoid......))))

There is such a thing. Still, if there are between 600-800 trades on the back and about 200 on the forward, then there is more confidence.

But there is no guarantee either.

 
LeoV >> :

I'm not arguing how you can see, I'm just asking -

I'm wondering - and that's fine. Normal question to your quote.

I've posted some stuff here, stuff that really works.There have also been hints/statements from other comrades on what to look out for.I think quite a few people here know about stuff that really works,just blatantly chasing stupidity.

 
FOXXXi писал(а) >>

I've posted some stuff here, stuff that really works.There have also been hints/statements from other comrades on what to look out for.I think quite a few people here know about stuff that really works,just blatantly chasing stupidity.

Yep, it all makes sense....))) Very instructive.....

 
FOXXXi >> :

... quite a few people here know what really works, they're just blatantly talking rubbish.

He who knows is silent, he who speaks does not know. (Lao Tzu)

 
LeoV >> :

I agree. I have come across that there is also an 'out-of-sample' fit. And this fit is harder to understand and avoid......))))

It is no longer a fit, but an unsteady market.


For example, Sample - sideways trends prevail, accordingly a counter-trend strategy will fit on it.

Out-Of-Sample - sideways trends continue. The strategy obtained on Samle will give a profit.


We put this strategy on the demo, the market changed to the trend direction. We made a loss. The strategy expired. Miracles don't happen. Eternal strategies do not exist either.


This has already happened many times, and there is nothing we can do about it.


The only consolation is that if the strategy expires, it starts losing with mathematical expectation minus spread per trade at constant lots (without MM). That is, if some other strategy has not expired and its expectation is much higher than spread, then it will not only profit, but also will cover losses of the "rotten" strategy.

 
Reshetov >> :

1)This is no longer a fit, but a non-stationary market.


For example, Sample - sideways trends prevail, accordingly a counter-trend strategy will pick up on it.

Out-Of-Sample - sideways trends continue. The strategy obtained on Samle gives a profit.


2) We put this strategy on the demo, and the market has changed in the trend direction. We took a loss. The strategy expired. Miracles do not happen. Eternal strategies also do not exist.


This has already happened many times, and nothing can be done.


3) The only consolation is that if the strategy expires, it starts losing with mathematical expectation minus Spread per trade at constant lots (without MM). That is, if some other strategy has not expired yet and its expectation is much higher than spread, then it will not only give profit, but also will cover losses of the "rotten" strategy.

What are you talking about? You're a smart guy who has been trading in financial markets for years.

1)No,it's the very real adjustment/optimization/training etc. to the non-stationarity of the market.

2) Don't say "the market has changed", it hasn't changed in any trend direction.

Eternal strategies can exist, or rather they work as long as the markets work.

3) All the strategies individually on martingales will add up additively, i.e. we will get several such strategies in parallel, together giving one such good losing strategy.

 
FOXXXi писал(а) >>

Here we go again with 25.And this is coming from you - a man who is not stupid and has been in the financial markets for years.Are you being silly? Are you still losing?

1)No, it's a real adjustment/optimization/training etc. to the non-stationarity of the market.

2) Don't say "the market has changed", it hasn't changed in any trend direction.

Eternal strategies can exist, or rather they work, as long as the markets work.

3) All the strategies one by one on martingales will additively, i.e. we obtain several such strategies together giving one such good losing strategy.

I am not against your statements. On the contrary, I'm even for it. But you have something more sensible to say, besides blaming everything? )))) Or if you can't, then can you at least confirm your words with some real life or monitoring? So that everyone can really get a feel for what you're saying. Not with screenshots. I don't need screenshots. ))))