Evaluating probabilistic forecasts with scoringRules

Jordan, Alexander; Krüger, Fabian; Lerch, Sebastian

doi:10.48550/arxiv.1709.04743

Cited by 14 publications

(15 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For every model we obtain empirical cumulative distribution functions of 10000 values for every hour. To obtain the CRPS we proceed like it is suggested in the R package scoringRules described in Jordan et al (2018). We compute the empirical quantiles with τ ∈ {0.001, .…”

Section: Evaluation Results -Univariate Copula-based Modelsmentioning

confidence: 99%

A copula-based time series model for global horizontal irradiation

Müller¹,

Reuber²

2020

Preprint

View full text Add to dashboard Cite

The increasing importance of solar power for electricity generation leads to an increasing demand for probabilistic forecasting of local and aggregated PV yields. In this paper we use an indirect modeling approach for hourly medium to long term local PV yields based on publicly available irradiation data. We suggest a time series model for global horizontal irradiation for which it is easy to generate an arbitrary number of scenarios and thus allows for multivariate probabilistic forecasts for arbitrary time horizons. In contrast to many simplified models that have been considered in the literature so far it features several important stylized facts. Sharp time dependent lower and upper bounds of global horizontal irradiations are estimated that improve the often used physical bounds. The parameters of the beta distributed marginals of the transformed data are allowed to be time dependent. A copula-based time series model is introduced for the hourly and daily dependence structure based on a simple graphical structure known from the theory of vine copulas. Non-Gaussian copulas like Gumbel and BB1 copulas are used that allow for the important feature of so-called tail dependence. Evaluation methods like the continuous ranked probability score (CRPS), the energy score (ES) and the variogram score (VS) are used to compare the power of the model for multivariate probabilistic forecasting with other models used in the literature showing that our model outperforms other models in many respects.

show abstract

Section: Evaluation Results -Univariate Copula-based Modelsmentioning

confidence: 99%

A copula-based time series model for global horizontal irradiation

Müller¹,

Reuber²

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Then the CRPS value at each time step can be attained from Eq. 13 according to [41]. In addition, CRPS sum can be obtained by summing across different dimensions and then averaged over the prediction horizon, that is…”

Section: A Datasets Description and Evaluation Metricsmentioning

confidence: 99%

ScoreGrad: Multivariate Probabilistic Time Series Forecasting with Continuous Energy-based Generative Models

Yan¹,

Zhang²,

Zhou³

et al. 2021

Preprint

View full text Add to dashboard Cite

Multivariate time series prediction has attracted a lot of attention because of its wide applications such as intelligence transportation, AIOps. Generative models have achieved impressive results in time series modeling because they can model data distribution and take noise into consideration. However, many existing works can not be widely used because of the constraints of functional form of generative models or the sensitivity to hyperparameters. In this paper, we propose ScoreGrad, a multivariate probabilistic time series forecasting framework based on continuous energy-based generative models. ScoreGrad is composed of time series feature extraction module and conditional stochastic differential equation based score matching module. The prediction can be achieved by iteratively solving reverse-time SDE. To the best of our knowledge, ScoreGrad is the first continuous energy based generative model used for time series forecasting. Furthermore, ScoreGrad achieves state-of-theart results on six real-world datasets. The impact of hyperparameters and sampler types on the performance are also explored.Code is available at https://github.com/yantijin/ScoreGradPred.

show abstract

“…To facilitate computation we assumed normality of the posterior distributions. We extracted means and standard deviations of the pointwise posteriors obtained from INLA and calculated the CRPS using the R package scoringRules (Jordan et al, 2017).…”

Section: Model Comparisonmentioning

confidence: 99%

Bayesian spatial modelling of terrestrial radiation in Switzerland

Folly¹,

Konstantinoudis²,

Mazzei-Abba³

et al. 2020

Preprint

View full text Add to dashboard Cite

The geographic variation of terrestrial radiation can be exploited in epidemiological studies of the health effects of protracted low-dose exposure. Various methods have been applied to derive maps of this variation. We aimed to construct a map of terrestrial radiation for Switzerland. We used airborne γ-spectrometry measurements to model the ambient dose rates from terrestrial radiation through a Bayesian mixed-effects model and conducted inference using Integrated Nested Laplace Approximation (INLA). We predicted higher levels of ambient dose rates in the alpine regions and Ticino compared with the western and northern parts of Switzerland. We provide a map that can be used for exposure assessment in epidemiological studies and as a baseline map for assessing potential contamination.

show abstract

Evaluating probabilistic forecasts with scoringRules

Cited by 14 publications

References 0 publications

A copula-based time series model for global horizontal irradiation

A copula-based time series model for global horizontal irradiation

ScoreGrad: Multivariate Probabilistic Time Series Forecasting with Continuous Energy-based Generative Models

Bayesian spatial modelling of terrestrial radiation in Switzerland

Contact Info

Product

Resources

About