Machine learning in trading: theory, models, practice and algo-trading - page 3564

New comment
 
Aleksey Vyazmikin #:

Am I correct, in the aggregate, that this is an implication that my definitions are made/thought out in a state of altered consciousness?

I'd like to understand the joke better.

In a state of flux :)

 
Maxim Dmitrievsky #:

In a state of flux :)

I think I agree :)

 

For genetics enthusiasts


 

That's an interesting thought

Если есть широко разрекламированные прогнозы о том, что обменный курс будет расти, то люди сразу же скорректируют цену, 
которую они готовы заплатить, и поэтому прогнозы сбываются. В каком-то смысле курсы валют становятся их собственными прогнозами.
 Это пример «гипотезы эффективного рынка». Следовательно, прогнозирование того, будет ли завтра обменный курс расти или падать,
 примерно так же предсказуемо, как и прогнозирование того, выпадет ли брошенная монета орлом или решкой


from the book

https://otexts.com/fpp3/what-can-be-forecast.html

 
mytarmailS #:

That's an interesting thought


from a book

https://otexts.com/fpp3/what-can-be-forecast.html

Recently sanctions were imposed - at the weekend exchangers raised the exchange rate, all the media shouted about the fall of the ruble, some people went to buy dollars, but on Monday the situation on Forex was different, i.e. the crowd's forecast did not come true.

So all this is chatter - some of the crowd's forecasts will come true, and some will not. And every smart person will find a good example for his words.

 
Aleksey Vyazmikin #:

Recently imposed sanctions - this weekend's exchanges.....

One case to pass off as a statistic? That's clever.
 
Aleksey Vyazmikin #:

Recently sanctions were imposed - at the weekend exchangers raised the exchange rate, all the media shouted about the fall of the ruble, some people went to buy dollars, but on Monday the situation on Forex was different, i.e. the forecast did not come true.

So all this is chatter - some of the crowd's forecasts will come true, and some will not. And every smart person will find a good example for his words.

Where did you see that?

Maybe you should change the media you read and widen your social circle ? :-)

 
mytarmailS #:
Passing off one case as a statistic? That's clever.

The smart thing to do would be to cite event statistics, but I don't have any. And it was quoted as 100% probability, which is why I wrote this post.

 
Maxim Kuznetsov #:

where did you even see that.

maybe you should change the media you read and widen your social circle ? :-)

Hmmm.... Well, look for yourself, what courses were for real money. Yes and now it has already diverged from the announced significantly ...

I will not continue the topic.

 
mytarmailS #:

That's an interesting thought

from a book

https://otexts.com/fpp3/what-can-be-forecast.html

I programmed a simple variant. some primitive simulation of the author's idea.

Arima starts a forecast in a sliding window with a piece of random series (first 100 points) and adds a new forecast point + noise at the end and the window shifts by +1. Eventually it turns out that the series consists only of forecasts and a new forecast is made only from the old forecasts.

We have got quite plausible series

library(forecast)

n <- 100
y <- rep(NA, 500)
y[1:n] <- cumsum(rnorm(n))


get_forec <- function(x){
  x |> auto.arima() |> forecast(h = 1) |> _$mean |> c()
}
for(i in n:(length(y)-1)){
  ii <- (i-99):i
  y[i+1] <- get_forec(y[ii])+rnorm(1)
  print(i)
}


plot(y, t="l")
abline(v=n,col=8,lty=2)
1...355735583559356035613562356335643565356635673568356935703571...3595
New comment