Econometrics: State-space model forecasting - page 24
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
it's funny...
it's funny...
you don't need it.
What you need to do -
1.first check the accuracy of simulation ... compare cuts obtained in software and realtime
2.look for connections.... you need a ruler... + select and examine influences... maybe you won't need Ensable18
ps.if you had posted the model on page 2 it would have been taken apart long ago...now I don't feel like poking around...Good Luck...
Well, there you go. There's your answer to your posts.
He doesn't want it -- he wants the swamp.
Don't lose the thread...
Sanych, come out of hiding... everyone here is human, not a beast and understands everything...
what you need to do -
You need to first verify the correctness of the simulation ... compare cuts obtained in the program and in real time
2.look for connections.... you need a ruler... + select and examine influences... maybe you won't need Ensemble 18
ps.if you had posted the model on page 2 it would have been taken apart long ago...now I don't feel like poking around...Good Luck...
I have stated my opinion above with advice to the author of the thread - get out of here. He needs to make money, not read the bullshit of the locals .... It's me who has nothing to do, so I'm stuck...
About downloads. From my profile so far:
wrapper downloads - 705.
indicator downloads - 1229
20 bars does not seem to me to be enough to calculate the condition.
In fact, fixing the window size is unlikely to do any good: critical information may well be dispersed in a much deeper history. I think this is the main drawback of autoregressive models in econometrics.
20 bars does not seem to me to be enough to calculate the condition.
In fact, fixing the window size is unlikely to do any good: critical information may well be dispersed in a much deeper history. In my opinion, this is the main drawback of autoregressive models in econometrics.
I use a dynamic adaptive state space model for non-stationary random processes. The choice of a particular model and its parameters is made by the arrival of each bar.
This model necessarily consists of at least two equations: the measurement equation, and the state equation(s). The state is predicted, and then a prediction of the measured quantity is calculated from this state prediction. There are variants of SSM that coincide with ARIMA, but this is a special and rather rare case that confirms your point.
Special kind of autoregression is used to calculate thresholds that are computed from prediction error, and it (prediction error) is stationary, i.e. autoregressive model is quite applicable to this random process.
As far as the window is concerned.
We need to answer the question: how many minimum history bars are needed for the trend to persist to the next bar? The probability of the trend persisting at 10+1 bars is much higher than at 50+1 bars and you may not consider 100+1 bars at all. Intuitively so.
I use a dynamic adaptive state-space model for non-stationary random processes. The choice of the particular model and its parameters is made by the arrival of each bar.
This model necessarily consists of at least two equations: the measurement equation, and the state equation(s). The state is predicted, and then a prediction of the measured quantity is calculated from this state prediction. There are variants of SSM that coincide with ARIMA, but this is a special and rather rare case that confirms your point.
Special kind of autoregression is used to calculate thresholds that are computed from prediction error, and it (prediction error) is stationary, i.e. autoregressive model is quite applicable to this random process.
As far as the window is concerned.
We need to answer the question: how many minimum history bars are needed for the trend to persist to the next bar? The probability of the trend persisting at 10+1 bars is much higher than at 50+1 bars and you may not consider 100+1 bars at all. Intuitively so.
I use a dynamic adaptive state-space model for non-stationary random processes. The choice of a particular model and its parameters is made by the arrival of each bar.
This model necessarily consists of at least two equations: the measurement equation, and the state equation(s). The state is predicted, and then a prediction of the measured quantity is calculated from this state prediction. There are variants of SSM that coincide with ARIMA, but this is a special and rather rare case that confirms your point.
Special kind of autoregression is used to calculate thresholds that are computed from prediction error, and it (prediction error) is stationary, i.e. autoregressive model is quite applicable to this random process.
As far as the window is concerned.
We need to answer the question: how many minimum history bars are needed for the trend to persist to the next bar? The probability of the trend persisting at 10+1 bars is much higher than at 50+1 bars and you may not consider 100+1 bars at all. Intuitively, it's like this.
1. Could you please quote the type of autoregression, the function on the basis of which the prediction is made.
2. I have to use up to 1000 bars of history, so cases >100 bars cannot be excluded. I should consider cases > 1000 bars, but for some reason the EA ignores these cases, although the indicator can display even 10000 bars. What is the reason in the Expert Advisor, I do not know. I cannot find the 1000-bar limit in the code. Perhaps this is a system constraint?