Econometrics: State-space model forecasting - page 15
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
The market is not stationary. So: either the model was originally conceived and really fits this non-stationarity or it doesn't. At me (I so conceived) it should correspond. Apart from that, the model has to fit the purpose. The goal is to follow the trend, and my model does not support this goal, but makes a point prediction. What is there to test? There is no object to be tested.
Well it matches the first screenshot... the trend has been told a hundred times... more is less from lag... I don't know why)))
How easy you make it... Especially the flip move. Snap, and you're in the hole. ;)))
The market is not stationary. Therefore: the model was originally designed and really corresponds to this non-stationarity, or not. In my case (I so conceived) it must comply. Apart from that, the model has to fit the purpose. The goal is to follow the trend, and my model does not support this goal, but makes a point prediction. What is there to test? There is no object to be tested.
Yeah. Looks like you're going to be all snarky and clever as long as you don't accidentally beat the tester... Then you'd have to go out into the real market... or admit you're an empty-headed idiot.... I'm about to shit my pants.
--
What are you talking about? :
The objective is to follow the trend, and my model does not support this objective, but makes a point forecast. What is there to test? There is no object to be tested.
I've explained to you in Russian and illustrated by code exactly how to convert "point forecast" to "trend forecast". Have you even read my previous post? Or took one quick glance and immediately rushed to refute it?
Artist, motherfucker. :))
It's so simple... Especially the flip move. Snap, and you're in the hole. ;)))
I'm a simple guy. I write simple. ;)
What's the big deal? It's all on the level playing field. You have to get fancy with forecasting, but with testing, you have to simplify things to the limit.
How easy you make it... Especially the flip move. Snap, and you're in the hole. ;)))
I'm a simple guy. I write simple. ;)
What's the big deal? It's all on the level playing field. You have to get fancy with forecasting, but with testing, you have to simplify things to the limit.
This "simplification to the limit in testing" leads to the pointlessness of "wisdom in prediction", and indeed of testing in general.
Yeah. Looks like you zvizdobolyat and clever until blue in the face, just so as not to accidentally beat the tester ... Then you'll have to go out into the real market... or admit you're an empty-headed idiot.... I'm about to shit my pants.
--
What are you talking about? :
I've explained to you in Russian and illustrated exactly how to convert "point forecast" to "trend forecast" by code. Did you even read my previous post? Or took one quick glance and immediately rushed to refute?
Artist, motherfucker. :))
To match the style of your post: utter stupidity, your suggestion.
Thanks to faa I found out that nobody has done it for 30 years in the way you suggest (first paper from 1983).
Such "simplification to the limit" leads to meaninglessness of testing in general.
In my experience so far it has not resulted in meaninglessness, but only in the rapid optimisation of "predictors" (if needed).
But I believe you of course. You know best of course. :)))
Thanks to faa, I found out that no one has done it for 30 years in the way you suggest (first paper from 1983)
"No one does it that way" is your strongest argument?
LOL.... ))))))
// Probably the same year in 1983 that the last econometrician managed to actually win a real $100 on the real exchange........ :-))