Journal
ENERGY SYSTEMS-OPTIMIZATION MODELING SIMULATION AND ECONOMIC ASPECTS
Volume 10, Issue 4, Pages 889-910Publisher
SPRINGER HEIDELBERG
DOI: 10.1007/s12667-018-0302-z
Keywords
Electricity consumption; Principal components regression; Vector autoregression; Turkey
Categories
Ask authors/readers for more resources
A first order vector autoregression topology was used to model and predict Turkey's net electricity consumption in the future. Input variables for the model were the annual values of electricity consumption along with four demographic and economic indicators such as, population, gross domestic product, imports and exports. Output variables were the one-step-ahead values of the same variables. First, polynomial regressions were used to determine and remove the trend components of all these five variables. Then, principal components regression methodwas applied to evaluate the coefficients of the vector autoregression model. Electricity consumption of Turkey was modeled using annual data from 1970 to 2016 and the model was used to predict future consumption values until year 2030. Singular value decomposition was used to determine the number of important dimensions in the data. This approach yielded a significant reduction in the dimensionality of the problem and thus provided robustness to the predictions. The results showed the feasibility of applying principal components regression method to vector autoregression model for electricity consumption prediction.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available