Where is the line between fitting and actual patterns? - page 20

 

Optimized for 2008. Tested outsample for 2009. Launched it for real in 2010.

If it leaks in 2011 - has it been tweaked or not?

 
paukas:

Optimized for 2008. Checked outsample for 2009. Launched it for real in 2010.

If it leaks in 2011 - was it a fit or not?


:-))) Try the following scheme - 1. 01.2006-06.2010 optimization, 06.10 - 22.01.2011 - autofsample - evaluate...

2. 01.2006 - 22.01.2011 - optimization, from 24.01.2011 - till 12.2011 - ripping foam on the real! :-)))

P.S. OOS no more than 25-30% of the optimization period.

 
Reshetov:
There is no need to repeat yourself. We already know your thoughts about the maliciousness of OOS tests by heart.
Yuri, where did I say that?
 
paukas:

Optimized for 2008. Checked Outsample for 2009. Ran it on real in 2010.

If he leaks in 2011 - was there a fit or not?

No, not if the OOS test was positive.

If there was a negative result on OOS, it does not necessarily mean a fit with Sample - the market can change. To be sure of the fit, it is necessary to conduct tests on OOS before Sample and after Sample. If both results are negative for both of these forwards, then we are dealing with a barefaced fitting.

Successful forward tests do not guarantee profitable TS in the future. Their purpose is to identify the fit.

 
lasso:
Yuri, where did I say that?

For the particularly gifted:

lasso:

...

My optimisation period is single and indivisible. ( joo and Avals have already written about the dangers of this division in the financial markets.)

...
 
Reshetov:

No, it wasn't, if the OOS test came back positive.


That's a strong consolation. :)
 
paukas:

Optimized for 2008. Tested outsample for 2009. Launched it for real in 2010.

If it leaks in 2011 - has it been tweaked or not?


You're going too fast, comrade )))) You can't go that fast for years ))))
 
LeoV:

You're going too far, comrade )))) You can't dash around like that for years ))))
And less than a year is no time at all.
 
Reshetov:

No, it wasn't, if the OOS test came back positive.

paukas:
That's a strong consolation. :)

Finally our opinions are the same, Vladimir.

Only you are just smiling and I almost collapsed.....

 

You guys are way off topic.

Don't listen to Reshetov about the AOC at all :) I'm speaking as a nerd with experience (senior nerd :) ).

Reason: