187 lines
11 KiB
BibTeX
187 lines
11 KiB
BibTeX
@article{berrisch2023modeling,
|
||
title = {Modeling volatility and dependence of European carbon and energy prices},
|
||
author = {Jonathan Berrisch and Sven Pappert and Florian Ziel and Antonia Arsova},
|
||
year = {2023},
|
||
month = {3},
|
||
journal = {Finance Research Letters},
|
||
publisher = {Elsevier BV},
|
||
volume = {52},
|
||
pages = {103503},
|
||
doi = {10.1016/j.frl.2022.103503},
|
||
issn = {1544-6123},
|
||
url = {https://arxiv.org/pdf/2208.14311},
|
||
abstract = {We study the prices of European Emission Allowances (EUA), whereby we analyze their uncertainty and dependencies on related energy prices (natural gas, coal, and oil). We propose a probabilistic multivariate conditional time series model with a VECM-Copula-GARCH structure which exploits key characteristics of the data. Data are normalized with respect to inflation and carbon emissions to allow for proper cross-series evaluation. The forecasting performance is evaluated in an extensive rolling-window forecasting study, covering eight years out-of-sample. We discuss our findings for both levels- and log-transformed data, focusing on time-varying correlations, and in view of the Russian invasion of Ukraine.},
|
||
keywords = {Carbon prices, Conditional volatility, Copula, Emission allowances, Energy markets, Forecasting, Multivariate modeling, Time series}
|
||
}
|
||
@article{marcjasz2022distributional,
|
||
title = {Distributional neural networks for electricity price forecasting},
|
||
author = {Marcjasz, Grzegorz and Narajewski, Micha{\l} and Weron, Rafa{\l} and Ziel, Florian},
|
||
journal = {Energy Economics},
|
||
volume = {125},
|
||
pages = {106843},
|
||
year = {2023},
|
||
doi = {10.1016/j.eneco.2023.106843},
|
||
publisher = {Elsevier}
|
||
}
|
||
@book{cesa2006prediction,
|
||
title = {{Prediction, learning, and games}},
|
||
author = {Cesa-Bianchi, Nicol{\`o} and Lugosi, G{\'a}bor},
|
||
year = {2006},
|
||
month = {3},
|
||
publisher = {Cambridge university press},
|
||
pages = {I--XII, 1--394},
|
||
doi = {10.1017/CBO9780511546921},
|
||
isbn = {978-0521841085}
|
||
}
|
||
@inproceedings{gaillard2018efficient,
|
||
title = {{Efficient online algorithms for fast-rate regret bounds under sparsity}},
|
||
author = {Gaillard, Pierre and Wintenberger, Olivier},
|
||
year = {2018},
|
||
month = {5},
|
||
journal = {arXiv (Cornell University)},
|
||
booktitle = {Proceedings of the 32nd International Conference on Neural Information Processing Systems},
|
||
publisher = {Cornell University},
|
||
pages = {7026--7036},
|
||
doi = {10.48550/arxiv.1805.09174},
|
||
url = {https://arxiv.org/abs/1805.09174},
|
||
editor = {Bengio, Samy and Wallach, Hanna M. and Larochelle, Hugo and Grauman, Kristen and Cesa-Bianchi, Nicolò and Garnett, Roman}
|
||
}
|
||
@article{gneiting2007strictly,
|
||
title = {{Strictly proper scoring rules, prediction, and estimation}},
|
||
author = {Gneiting, Tilmann and Raftery, Adrian E},
|
||
year = {2007},
|
||
month = {3},
|
||
journal = {Journal of the American statistical Association},
|
||
publisher = {Taylor \& Francis},
|
||
volume = {102},
|
||
number = {477},
|
||
pages = {359--378},
|
||
doi = {10.1198/016214506000001437},
|
||
issn = {0162-1459}
|
||
}
|
||
@article{gneiting2011making,
|
||
title = {{Making and evaluating point forecasts}},
|
||
author = {Gneiting, Tilmann},
|
||
year = {2011},
|
||
month = {6},
|
||
journal = {Journal of the American Statistical Association},
|
||
publisher = {Taylor \& Francis},
|
||
volume = {106},
|
||
number = {494},
|
||
pages = {746--762},
|
||
doi = {10.1198/jasa.2011.r10138},
|
||
issn = {0162-1459},
|
||
url = {https://arxiv.org/pdf/0912.0902.pdf}
|
||
}
|
||
@article{wintenberger2017optimal,
|
||
title = {Optimal learning with Bernstein online aggregation},
|
||
author = {Wintenberger, Olivier},
|
||
year = {2017},
|
||
journal = {Machine Learning},
|
||
publisher = {Springer},
|
||
volume = {106},
|
||
number = {1},
|
||
pages = {119--141},
|
||
doi = {10.1007/s10994-016-5592-6}
|
||
}
|
||
@article{BERRISCH2023105221,
|
||
title = {CRPS learning},
|
||
author = {Jonathan Berrisch and Florian Ziel},
|
||
year = {2023},
|
||
month = {12},
|
||
journal = {Journal of Econometrics},
|
||
publisher = {Elsevier BV},
|
||
volume = {237},
|
||
number = {2, Part C},
|
||
pages = {105221},
|
||
doi = {10.1016/j.jeconom.2021.11.008},
|
||
issn = {0304-4076},
|
||
url = {https://arxiv.org/pdf/2102.00968},
|
||
abstract = {Combination and aggregation techniques can significantly improve forecast accuracy. This also holds for probabilistic forecasting methods where predictive distributions are combined. There are several time-varying and adaptive weighting schemes such as Bayesian model averaging (BMA). However, the quality of different forecasts may vary not only over time but also within the distribution. For example, some distribution forecasts may be more accurate in the center of the distributions, while others are better at predicting the tails. Therefore, we introduce a new weighting method that considers the differences in performance over time and within the distribution. We discuss pointwise combination algorithms based on aggregation across quantiles that optimize with respect to the continuous ranked probability score (CRPS). After analyzing the theoretical properties of pointwise CRPS learning, we discuss B- and P-Spline-based estimation techniques for batch and online learning, based on quantile regression and prediction with expert advice. We prove that the proposed fully adaptive Bernstein online aggregation (BOA) method for pointwise CRPS online learning has optimal convergence properties. They are confirmed in simulations and a probabilistic forecasting study for European emission allowance (EUA) prices.},
|
||
keywords = {Combination, Aggregation, Online, Probabilistic, Forecasting, Quantile, Time series, Distribution, Density, Prediction, Splines}
|
||
}
|
||
|
||
@article{BERRISCH20241568,
|
||
title = {Multivariate probabilistic CRPS learning with an application to day-ahead electricity prices},
|
||
author = {Jonathan Berrisch and Florian Ziel},
|
||
year = {2024},
|
||
month = {10},
|
||
journal = {International Journal of Forecasting},
|
||
publisher = {Elsevier BV},
|
||
volume = {40},
|
||
number = {4},
|
||
pages = {1568--1586},
|
||
doi = {10.1016/j.ijforecast.2024.01.005},
|
||
issn = {0169-2070},
|
||
abstract = {This paper presents a new method for combining (or aggregating or ensembling) multivariate probabilistic forecasts, considering dependencies between quantiles and marginals through a smoothing procedure that allows for online learning. We discuss two smoothing methods: dimensionality reduction using Basis matrices and penalized smoothing. The new online learning algorithm generalizes the standard CRPS learning framework into multivariate dimensions. It is based on Bernstein Online Aggregation (BOA) and yields optimal asymptotic learning properties. The procedure uses horizontal aggregation, i.e., aggregation across quantiles. We provide an in-depth discussion on possible extensions of the algorithm and several nested cases related to the existing literature on online forecast combination. We apply the proposed methodology to forecasting day-ahead electricity prices, which are 24-dimensional distributional forecasts. The proposed method yields significant improvements over uniform combination in terms of continuous ranked probability score (CRPS). We discuss the temporal evolution of the weights and hyperparameters and present the results of reduced versions of the preferred model. A fast C＋＋implementation of the proposed algorithm is provided in the open-source R-Package profoc on CRAN.},
|
||
keywords = {Combination, Aggregation, Ensembling, Online, Multivariate, Probabilistic, Forecasting, Quantile, Time series, Distribution, Density, Prediction, Splines}
|
||
}
|
||
@article{berrisch2025rcpptimer,
|
||
title = {rcpptimer: Rcpp Tic-Toc Timer with OpenMP Support},
|
||
author = {Berrisch, Jonathan},
|
||
year = {2025},
|
||
month = {1},
|
||
journal = {arXiv preprint arXiv:2501.15856},
|
||
publisher = {Cornell University},
|
||
volume = {abs/2501.15856},
|
||
doi = {10.48550/arXiv.2501.15856},
|
||
issn = {2331-8422},
|
||
url = {https://arxiv.org/pdf/2501.15856}
|
||
}
|
||
@article{hirsch2024online,
|
||
title = {Online Distributional Regression},
|
||
author = {Hirsch, Simon and Berrisch, Jonathan and Ziel, Florian},
|
||
year = {2024},
|
||
month = {6},
|
||
journal = {arXiv preprint arXiv:2407.08750},
|
||
publisher = {Cornell University},
|
||
volume = {abs/2407.08750},
|
||
doi = {10.48550/arXiv.2407.08750},
|
||
issn = {2331-8422},
|
||
url = {https://arxiv.org/pdf/2407.08750}
|
||
}
|
||
@article{berrisch2022distributional,
|
||
title = {Distributional modeling and forecasting of natural gas prices},
|
||
author = {Berrisch, Jonathan and Ziel, Florian},
|
||
year = {2022},
|
||
month = {1},
|
||
journal = {Journal of Forecasting},
|
||
publisher = {Wiley},
|
||
volume = {41},
|
||
number = {6},
|
||
pages = {1065--1086},
|
||
doi = {10.1002/for.2853},
|
||
issn = {0277-6693},
|
||
url = {https://arxiv.org/pdf/2010.06227},
|
||
abstract = {Abstract We examine the problem of modeling and forecasting European day-ahead and month-ahead natural gas prices. For this, we propose two distinct probabilistic models that can be utilized in risk and portfolio management. We use daily pricing data ranging from 2011 to 2020. Extensive descriptive data analysis shows that both time series feature heavy tails and conditional heteroscedasticity and show asymmetric behavior in their differences. We propose state-space time series models under skewed, heavy-tailed distributions to capture all stylized facts of the data. They include the impact of autocorrelation, seasonality, risk premia, temperature, storage levels, the price of European Emission Allowances, and related fuel prices of oil, coal, and electricity. We provide rigorous model diagnostics and interpret all model components in detail. Additionally, we conduct a probabilistic forecasting study with significance tests and compare the predictive performance against literature benchmarks. The proposed day-ahead (month-ahead) model leads to a 13\% (9\%) reduction in out-of-sample continuous ranked probability score (CRPS) compared with the best performing benchmark model, mainly due to adequate modeling of the volatility and heavy tails.},
|
||
keywords = {gas prices, heavy-tailed distribution, probabilistic forecasting, risk premium, state-space models}
|
||
}
|
||
@article{BERRISCH2023100236,
|
||
title = {High-resolution peak demand estimation using generalized additive models and deep neural networks},
|
||
author = {Jonathan Berrisch and Micha{\l} Narajewski and Florian Ziel},
|
||
year = {2023},
|
||
month = {7},
|
||
journal = {Energy and AI},
|
||
publisher = {Elsevier BV},
|
||
volume = {13},
|
||
pages = {100236},
|
||
doi = {10.1016/j.egyai.2023.100236},
|
||
issn = {2666-5468},
|
||
keywords = {Electricity peak load, Generalized additive models, Artificial neural networks, Prediction, Combination, Weather effects, Seasonality}
|
||
}
|
||
@book{johnson1995continuous,
|
||
title = {Continuous univariate distributions, volume 2},
|
||
author = {Johnson, Norman L and Kotz, Samuel and Balakrishnan, Narayanaswamy},
|
||
year = {1995},
|
||
publisher = {John wiley \& sons},
|
||
volume = {289}
|
||
}
|
||
@article{li2022general,
|
||
title = {General P-splines for non-uniform B-splines},
|
||
author = {Li, Zheyuan and Cao, Jiguo},
|
||
year = {2022},
|
||
journal = {arXiv preprint},
|
||
publisher = {Cornell University},
|
||
doi = {10.48550/arXiv.2201.06808},
|
||
url = {https://arxiv.org/abs/2201.06808}
|
||
} |