Forecasting indicators - page 32

[Mladen Rakic]

krelian99:
It is only for offline data analysis since it repaints.

Just a clarification : averages part is not repainting. Extrapolation part (the part forecasting the future) is, like any extrapolation, recalculating/repainting

[spacejockey]

mladen:
Just a clarification : averages part is not repainting. Extrapolation part (the part forecasting the future) is, like any extrapolation, recalculating/repainting

Hi guys

thx a lot, I am beginning to learn Demark TD Theory , I know some  Forecasting Theory, as Gann Theory,DELTA Theory

I want to ask  are there other Forecasting theories ?

Regards. 

[yenPound]

blueface:

Hi guys

thx a lot, I am beginning to learn Demark TD Theory , I know some  Forecasting Theory, as Gann Theory,DELTA Theory

I want to ask  are there other Forecasting theories ?

Regards. 

Causal / econometric forecasting methods

• Regression analysis includes a large group of methods for predicting future values of a variable using information about other variables. These methods include both parametric (linear or non-linear) and non-parametric techniques.
• Autoregressive moving average with exogenous inputs (ARMAX)^[4]

_____________________________________________________________________________________________

Artificial intelligence methods

• Artificial neural networks
• Group method of data handling
• Support vector machines

Often these are done today by specialized programs loosely labeled

• Data mining
• Machine Learning
• Pattern Recognition

Time series methods

Time series methods use historical data as the basis of estimating future outcomes.

• Moving average
• Weighted moving average
• Kalman filtering
• Exponential smoothing
• Autoregressive moving average (ARMA)
• Autoregressive integrated moving average (ARIMA)

e.g. Box–JenkinsSeasonal ARIMA or SARIMA

• Extrapolation
• Linear prediction
• Trend estimation
• Growth curve (statistics)

[myname]

KumoBreake:

Causal / econometric forecasting methods

• Regression analysis includes a large group of methods for predicting future values of a variable using information about other variables. These methods include both parametric (linear or non-linear) and non-parametric techniques.
• Autoregressive moving average with exogenous inputs (ARMAX)^[4]

_____________________________________________________________________________________________

Artificial intelligence methods

• Artificial neural networks
• Group method of data handling
• Support vector machines

Often these are done today by specialized programs loosely labeled

• Data mining
• Machine Learning
• Pattern Recognition

Time series methods

Time series methods use historical data as the basis of estimating future outcomes.

• Moving average
• Weighted moving average
• Kalman filtering
• Exponential smoothing
• Autoregressive moving average (ARMA)
• Autoregressive integrated moving average (ARIMA)

e.g. Box–JenkinsSeasonal ARIMA or SARIMA

• Extrapolation
• Linear prediction
• Trend estimation
• Growth curve (statistics)

Nice. Where it came from?

[yenPound]

nbtrading:
Nice. Where it came from?

https://en.wikipedia.org/wiki/Forecasting  

:))

[myname]

KumoBreake:

https://en.wikipedia.org/wiki/Forecasting  

:))

And I thought that you were compiling the links :)

[-IXI-]

mladen:

Fourier extrapolation of some of the averages (20 types in this version) :

ATTENTION: Video should be reuploaded

fourier_extrapolation_of_averages-2_1.mq4

_________________

PS: "Fourier extrapolation.dll" must be in the libraries folder

ATTENTION: Video should be reuploaded

fourier_extrapolation.zip

Dear mladen

Is there anyway to call future value of the Fourier Extrapolation for testing in an EA? Future EA?!

Best,

[Mladen Rakic]

IXI:

Dear mladen

Is there anyway to call future value of the Fourier Extrapolation for testing in an EA? Future EA?!

Best,

If you check the bar 0, which is normally the current bar, for Fourier extrapolation that is the rightmost - the "most in the future" value. And so on. You have to start checking the bars closer to "future bars" value to be closer to the present. So, all in all, you are always having future values when reading extrapolation buffer values from bar 0 to bar "future bars"-1 which is the current bar in that case

[myname]

IXI:

Dear mladen

Is there anyway to call future value of the Fourier Extrapolation for testing in an EA? Future EA?!

Best,

Wouldn't that be too risky?

[Mladen Rakic]

IXI:

Dear mladen

Thanks for comments, but I have not succeeded to test the indicator in EA. I tested to call difference between moving average buffer and future buffer or compare difference between future buffer values for example 0 and -1 or -2 with no success. In addition I noticed if you change future bar number input the result is always same and number of predicted bars can not be adjusted. Any comment?

nbtrading

I just want to check the future prediction result in EA with a normal moving average. A normal moving average if you open and close trades with its signal has a success rate between 20 to 40 percent, usually around 30 percent. Is  result from extrapolation which predict future is different? Maybe.

Best,

You can not use the value -1 or -2 for the buffer element. You simply have to use +1 and +2 and then you shall get the values that are first to the latest prediction (the one "most in the future")