That's interesting - page 23

 

HideYourRichess:

"The degree of completeness of the model may vary, but everything must be 'physical'. Behind each parameter there must be a real phenomenon." ... starting with the question "what is your process?
Give any example (not necessarily markets) of an approach to model building that is adequate from your point of view.
 
Farnsworth:


... And colleagues are in the way, there, Mishek claims the branch, grumbling :o)))

It's not me claiming the branch, it's you who don't know what to do with it. The thread with the title "This is interesting" and Schmoll on the horse, had a chance with HideYourRichess, at best, to be transformed into "Exchange of research methodologies".

 
HideYourRichess:

...

. Indeed. We're done.

. Very good. Other than the annoyance of having to think about crowd psychology and shit, I don't feel anything else. Like, let's add something to something, divide it, subtract the "bottom" from the "top" and get a Jedi mind channel. I'm sick of this nonsense. The directions you have outlined are for petricks, will make good money on books, courses, techniques.

PS:

.Yes, we are wrapping up. But all the same, it was nice talking to you. Good luck.

.(:о)

 
Mischek:

It's not me claiming the branch, it's you who don't know what to do with it. The thread with the title "This is interesting" and Schmoll on the horse, had a chance with HideYourRichess, at best, to be transformed into "Exchange of research methodologies".


Why so soon - I don't know? Everything I wanted to do in the first post - I've already done, just for the sake of familiarity. And HideYourRichess said nothing new. I started sniffing around forex and stocks back in 90. I was a TA fan and lived happily ever after until one day I realised that I was blindfolded and walking on the edge of a precipice.

Yes, you can trade for a relatively long time using TA while analyzing all available information, but the formula for such successful trading: "information + analysis + intuition + intuition". I have a need for an automated system that will not lead to success. And what I need is ATC, didn't I say?
 
Farnsworth:

.Very good. Other than annoyance at the idea of crowd psychology and all that crap, I don't feel much else. Like, let's add something to something, divide it, subtract the "bottom" from the "top" and get a channel of Jedi power. I'm sick of this nonsense. The directions you have outlined are for petricks, will make good money on books, courses, methodologies.

PS:

.Yes, we are wrapping up. But all the same, it was nice talking to you. Good luck.

.(:о)

. What's there to say... Our exchange of views is a good example of how different the ideas on how to solve the common problem of taking money out of the market can be. It goes so far as not understanding each other with the same words. That's why I stand by my opinion, - what you suggest is a simple mechanistic application of abstract mathematical formulas to the market, without any attempt to reflect the essence of the processes and phenomena taking place.

. By the way, if your sense of humour and beauty is still not completely lost, here's a present for you. :) I've invented a caption to your drawing: an illustrative example of martingale-deposit obtained by long and complex mathematical transformations from martingale-price.

. All the best and good luck.

 
hrenfx:
Give any example (not necessarily markets) of an approach to model building that you think is adequate.
An intersection model, a link capacity model, a free fall acceleration formula, simple arithmetic, etc. In the end - a model with long legs, a very adequate model.
 
HideYourRichess:
A junction model, a bandwidth model, a free fall acceleration formula, simple arithmetic, etc. In the end - a model with long legs, a very adequate model.

Why call GARCH a model?
 
HideYourRichess:

. What's there to say... Our exchange of views is a good example of how different the ideas on how to solve the common problem of taking money out of the market can be. It goes so far as not understanding each other with the same words. That's why I stand by my opinion, - what you suggest is a simple mechanistic application of abstract mathematical formulas to the market, without any attempt to reflect the essence of the processes and phenomena taking place.

. By the way, if your sense of humour and beauty is still not completely lost, here's a present for you. :) I've invented a caption to your drawing: an illustrative example of martingale-deposit obtained by long and complex mathematical transformations from martingale-price.

. Best of luck and good luck.


TA - doesn't capture any essence of the phenomenon, none at all. But you are absolutely right that it must be done, otherwise it is impossible to get away from martingale in its various manifestations and fundamentally impossible to build a strategy to which you can entrust your deposit.

That is exactly the kind of system I am trying to build. Gradually, it has 7 components according to the blueprints, I have told you about one so far.

PS: the "martingale deposit" has a statistically proven/confirmed trend, and the rest of the plots. This is important. And remember - price martingale is not complete.

 
Farnsworth:

TA - does not capture any essence of the phenomenon, none at all. But you are absolutely right that it has to be done, otherwise it is impossible to get away from martingale in its various manifestations and fundamentally impossible to build a strategy to which you can entrust your deposit.

That is exactly the kind of system I am trying to build. Gradually, according to the blueprints there are 7 components in it, I have told you about one so far.

If the CwRs are martingales - then you can't squeeze anything out of them. Martingales imply the absence of any pattern. No matter how you spin a martingale, you get a martingale.

From my point of view, the RDCs are not martingales because some mixing of some RDCs gives BPs with very stable values at Out of Sample in length comparable to the length of analysis for mixing the original RDCs.

 
hrenfx:

Why do they call GARCH a model?

. Because this is a common mistake. GARCH is just a mathematical method. This method, applied with care and diligence to a real process can become a model if it reflects the essence of the process itself. Otherwise, you get a situation of "pulling the market on formulas" (someone on the spider), and the market will pull on you in retaliation. Which, in fact, we see all the time.

. I'm sick and tired of repeating the same commonplace truths.