Difference between revisions of "Orange: Model Evaluation"

From OnnoWiki
Jump to navigation Jump to search
(Created page with "Sumber: https://orange.biolab.si/widget-catalog/time-series/model_evaluation/ Evaluate different time series’ models. Inputs Time series: Time series as output by As...")
 
 
(5 intermediate revisions by the same user not shown)
Line 2: Line 2:
  
  
−
Evaluate different time series’ models.
+
Widget Model Evaluation meng-evaluasi berbagai model time series
  
−
Inputs
+
==Input==
  
−
    Time series: Time series as output by As Timeseries widget.
+
Time series: Time series as output by As Timeseries widget.
−
    Time series model(s): The time series model(s) to evaluate (e.g. VAR or ARIMA).
+
Time series model(s): The time series model(s) to evaluate (e.g. VAR or ARIMA).
  
−
Evaluate different time series’ models. by comparing the errors they make in terms of: root mean squared error (RMSE), median absolute error (MAE), mean absolute percent error (MAPE), prediction of change in direction (POCID), coefficient of determination (R²), Akaike information criterion (AIC), and Bayesian information criterion (BIC).
+
Widget Model Evaluation dapat mengevaluasi berbagai model time series. Dengan membandingkan error yang di dihasilkan, yaitu:
  
−
    Number of folds for time series cross-validation.
+
* root mean squared error (RMSE)
−
    Number of forecast steps to produce in each fold.
+
* median absolute error (MAE)
−
    Results for various error measures and information criteria on cross-validated and in-sample data.
+
* mean absolute percent error (MAPE)
 +
* prediction of change in direction (POCID)
 +
* coefficient of determination (R²)
 +
* Akaike information criterion (AIC)
 +
* Bayesian information criterion (BIC).
  
−
This slide (source) shows how cross validation on time series is performed. In this case, the number of folds (1) is 10 and the number of forecast steps in each fold (2) is 1.
+
[[File:Model-evaluation-stamped.png|center|600px|thumb]]
  
−
In-sample errors are the errors calculated on the training data itself. A stable model is one where in-sample errors and out-of-sample errors don’t differ significantly.
+
* Number of folds for time series cross-validation.
 +
* Number of forecast steps to produce in each fold.
 +
* Results for various error measures and information criteria on cross-validated and in-sample data.
  
−
####See also
+
[[File:Automatic-time-series-forecasting-71-638.jpg|center|200px|thumb]]
 +
 
 +
Slide https://www.slideshare.net/hyndman/automatic-time-series-forecasting memperlihatkan bagaimana cara melakukan cross validation pada time series dilakukan. Dalam hal ini, jumlah dari folds (1) adalah 10 dan jumlah dari forecast steps di setiap fold (2) adalah 1.
 +
 
 +
In-sample errors adalah error yang di hitung pada training data itu sendiri. Sebuah model yang stabil adalah dimana in-sample errors dan out-of-sample errors tidak berbeda terlalu jauh.
 +
 
 +
==See also==
  
 
ARIMA Model, VAR Model
 
ARIMA Model, VAR Model

Latest revision as of 09:32, 7 April 2020

Sumber: https://orange.biolab.si/widget-catalog/time-series/model_evaluation/


Widget Model Evaluation meng-evaluasi berbagai model time series

Input

Time series: Time series as output by As Timeseries widget.
Time series model(s): The time series model(s) to evaluate (e.g. VAR or ARIMA).

Widget Model Evaluation dapat mengevaluasi berbagai model time series. Dengan membandingkan error yang di dihasilkan, yaitu:

  • root mean squared error (RMSE)
  • median absolute error (MAE)
  • mean absolute percent error (MAPE)
  • prediction of change in direction (POCID)
  • coefficient of determination (R²)
  • Akaike information criterion (AIC)
  • Bayesian information criterion (BIC).
Model-evaluation-stamped.png
  • Number of folds for time series cross-validation.
  • Number of forecast steps to produce in each fold.
  • Results for various error measures and information criteria on cross-validated and in-sample data.
Automatic-time-series-forecasting-71-638.jpg

Slide https://www.slideshare.net/hyndman/automatic-time-series-forecasting memperlihatkan bagaimana cara melakukan cross validation pada time series dilakukan. Dalam hal ini, jumlah dari folds (1) adalah 10 dan jumlah dari forecast steps di setiap fold (2) adalah 1.

In-sample errors adalah error yang di hitung pada training data itu sendiri. Sebuah model yang stabil adalah dimana in-sample errors dan out-of-sample errors tidak berbeda terlalu jauh.

See also

ARIMA Model, VAR Model


Referensi

Pranala Menarik

Retrieved from "https://onnocenter.or.id/wiki/index.php?title=Orange:_Model_Evaluation&oldid=60951"