PCA Forecast Averaging—Predicting Day-Ahead and Intraday Electricity Prices
Recently, the development in combining point forecasts of electricity prices obtained with different length of calibration windows have provided an extremely efficient and simple tool for improving predictive accuracy. However, the proposed methods are strongly dependent on expert knowledge and may...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-07-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/13/14/3530 |
| _version_ | 1797562893338673152 |
|---|---|
| author | Katarzyna Maciejowska Bartosz Uniejewski Tomasz Serafin |
| author_facet | Katarzyna Maciejowska Bartosz Uniejewski Tomasz Serafin |
| author_sort | Katarzyna Maciejowska |
| collection | DOAJ |
| description | Recently, the development in combining point forecasts of electricity prices obtained with different length of calibration windows have provided an extremely efficient and simple tool for improving predictive accuracy. However, the proposed methods are strongly dependent on expert knowledge and may not be directly transferred from one to another model or market. Hence, we consider a novel extension and propose to use principal component analysis (PCA) to automate the procedure of averaging over a rich pool of predictions. We apply PCA to a panel of over 650 point forecasts obtained for different calibration windows length. The robustness of the approach is evaluated with three different forecasting tasks, i.e., forecasting day-ahead prices, forecasting intraday ID3 prices one day in advance, and finally very short term forecasting of ID3 prices (i.e., six hours before delivery). The empirical results are compared using the Mean Absolute Error measure and Giacomini and White test for conditional predictive ability (CPA). The results indicate that PCA averaging not only yields significantly more accurate forecasts than individual predictions but also outperforms other forecast averaging schemes. |
| first_indexed | 2024-03-10T18:36:15Z |
| format | Article |
| id | doaj.art-189d4886967c457db3f7948550d5ab2b |
| institution | Directory Open Access Journal |
| issn | 1996-1073 |
| language | English |
| last_indexed | 2024-03-10T18:36:15Z |
| publishDate | 2020-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj.art-189d4886967c457db3f7948550d5ab2b2023-11-20T06:13:55ZengMDPI AGEnergies1996-10732020-07-011314353010.3390/en13143530PCA Forecast Averaging—Predicting Day-Ahead and Intraday Electricity PricesKatarzyna Maciejowska0Bartosz Uniejewski1Tomasz Serafin2Department of Operations Research and Business Intelligence, Wrocław University of Science and Technology, 50-370 Wrocław, PolandDepartment of Operations Research and Business Intelligence, Wrocław University of Science and Technology, 50-370 Wrocław, PolandDepartment of Operations Research and Business Intelligence, Wrocław University of Science and Technology, 50-370 Wrocław, PolandRecently, the development in combining point forecasts of electricity prices obtained with different length of calibration windows have provided an extremely efficient and simple tool for improving predictive accuracy. However, the proposed methods are strongly dependent on expert knowledge and may not be directly transferred from one to another model or market. Hence, we consider a novel extension and propose to use principal component analysis (PCA) to automate the procedure of averaging over a rich pool of predictions. We apply PCA to a panel of over 650 point forecasts obtained for different calibration windows length. The robustness of the approach is evaluated with three different forecasting tasks, i.e., forecasting day-ahead prices, forecasting intraday ID3 prices one day in advance, and finally very short term forecasting of ID3 prices (i.e., six hours before delivery). The empirical results are compared using the Mean Absolute Error measure and Giacomini and White test for conditional predictive ability (CPA). The results indicate that PCA averaging not only yields significantly more accurate forecasts than individual predictions but also outperforms other forecast averaging schemes.https://www.mdpi.com/1996-1073/13/14/3530electricity price forecastingEPFday-ahead marketintraday marketforecast averagingprincipal component analysis |
| spellingShingle | Katarzyna Maciejowska Bartosz Uniejewski Tomasz Serafin PCA Forecast Averaging—Predicting Day-Ahead and Intraday Electricity Prices Energies electricity price forecasting EPF day-ahead market intraday market forecast averaging principal component analysis |
| title | PCA Forecast Averaging—Predicting Day-Ahead and Intraday Electricity Prices |
| title_full | PCA Forecast Averaging—Predicting Day-Ahead and Intraday Electricity Prices |
| title_fullStr | PCA Forecast Averaging—Predicting Day-Ahead and Intraday Electricity Prices |
| title_full_unstemmed | PCA Forecast Averaging—Predicting Day-Ahead and Intraday Electricity Prices |
| title_short | PCA Forecast Averaging—Predicting Day-Ahead and Intraday Electricity Prices |
| title_sort | pca forecast averaging predicting day ahead and intraday electricity prices |
| topic | electricity price forecasting EPF day-ahead market intraday market forecast averaging principal component analysis |
| url | https://www.mdpi.com/1996-1073/13/14/3530 |
| work_keys_str_mv | AT katarzynamaciejowska pcaforecastaveragingpredictingdayaheadandintradayelectricityprices AT bartoszuniejewski pcaforecastaveragingpredictingdayaheadandintradayelectricityprices AT tomaszserafin pcaforecastaveragingpredictingdayaheadandintradayelectricityprices |