| Title: | Weighted Ensemble for Hybrid Model |
|---|---|
| Description: | The weighted ensemble method is a valuable approach for combining forecasts. This algorithm employs several optimization techniques to generate optimized weights. This package has been developed using algorithm of Armstrong (1989) <doi:10.1016/0024-6301(90)90317-W>. |
| Authors: | Dr. Ranjit Kumar Paul [aut], Dr. Md Yeasin [aut, cre] |
| Maintainer: | Dr. Md Yeasin <[email protected]> |
| License: | GPL-3 |
| Version: | 0.1.0 |
| Built: | 2025-02-18 03:39:58 UTC |
| Source: | https://github.com/cran/WeightedEnsemble |
Weighted Ensemble for Hybrid Model
WeightedEnsemble(df, Method = "PSO", test_data = NULL, forecast = NULL)WeightedEnsemble(df, Method = "PSO", test_data = NULL, forecast = NULL)
df |
Data set (training result) with first column as observed value |
Method |
Method of optimization |
test_data |
Test result |
forecast |
Forecast result |
Weights: Optimized weight
Optimized_Result: Optimized result
J. S. Armstrong. Combining forecasts: The end of the beginning or the beginning of the end? International Journal of Forecasting, 5(4):585–588, 1989.
y1<-rnorm(100,mean=100,sd=50) y2<- rnorm(100,mean=100,sd=50) y3<- rnorm(100,mean=100,sd=50) y4<-rnorm(100,mean=100,sd=50) y<-rnorm(100,mean=100,sd=50) data<-cbind(y,y1,y2,y3,y4) OptiSemble<-WeightedEnsemble(df=data)y1<-rnorm(100,mean=100,sd=50) y2<- rnorm(100,mean=100,sd=50) y3<- rnorm(100,mean=100,sd=50) y4<-rnorm(100,mean=100,sd=50) y<-rnorm(100,mean=100,sd=50) data<-cbind(y,y1,y2,y3,y4) OptiSemble<-WeightedEnsemble(df=data)