| Type: | Package |
| Title: | Weighted Ensemble for Hybrid Model |
| Version: | 0.1.0 |
| Author: | Dr. Ranjit Kumar Paul [aut], Dr. Md Yeasin [aut, cre] |
| Maintainer: | Dr. Md Yeasin <yeasin.iasri@gmail.com> |
| 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>. |
| License: | GPL-3 |
| Encoding: | UTF-8 |
| Imports: | stats, metaheuristicOpt |
| RoxygenNote: | 7.2.1 |
| NeedsCompilation: | no |
| Packaged: | 2023-04-07 17:41:25 UTC; YEASIN |
| Repository: | CRAN |
| Date/Publication: | 2023-04-10 14:10:06 UTC |
Weighted Ensemble for Hybrid Model
Description
Weighted Ensemble for Hybrid Model
Usage
WeightedEnsemble(df, Method = "PSO", test_data = NULL, forecast = NULL)
Arguments
df |
Data set (training result) with first column as observed value |
Method |
Method of optimization |
test_data |
Test result |
forecast |
Forecast result |
Value
Weights: Optimized weight
Optimized_Result: Optimized result
References
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.
Examples
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)