Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,657 papers · 148 categories

Trend · papers per month

2468 · Jun 202619922001200920172026
48 results for CRPS

Kernel quadrature improves CRPS estimation for probabilistic time-series forecasting.

problem Intractable integrations in CRPS evaluation metrics lead to improper rankings of forecasting models.
method Introduced kernel quadrature approach for unbiased CRPS estimation and scalable computation.
result Our approach consistently outperforms existing CRPS estimators.

Study improves distributional regression evaluation with CRPS, finding optimal rates of convergence.

problem Improving probabilistic forecasts in meteorology using distributional regression.
method Extends theoretical properties of CRPS evaluation to include covariates and finite sample sizes, analyzing convergence rates for different methods.
result Optimal minimax rate of convergence for distributional regression methods is achieved by k-nearest neighbor and kernel methods.

A method for non-parametric conditional distribution estimation using CRPS-optimal binning.

problem Non-parametric conditional distribution estimation.
method Partitioning covariate-sorted observations into bins to minimize LOO-CRPS, selecting K by K-fold cross-validation of test CRPS.
result Produces narrower prediction intervals with near-nominal coverage compared to split-conformal competitors.

Develop a framework to evaluate the reliability of probabilistic emulation of physical systems.

problem Evaluating the reliability of probabilistic forecasts in physical systems.
method Developing a framework to assess the reliability of probabilistic emulation across diverse 2D spatiotemporal systems.
result CRPS-trained ensembles achieve more reliable uncertainties on single-step prediction and autoregressive rollouts.

Study finds CRPS learning doesn't improve day-ahead bidding profits despite better accuracy.

problem Improving day-ahead bidding profits through better probabilistic price forecasting.
method CRPS learning to minimize continuous ranked probability score (CRPS) for ensemble predictions.
result Higher diversity in ensemble predictions improves accuracy but doesn't lead to higher profits.

New method improves forecast accuracy using CRPS for probabilistic predictions.

problem Improving forecast accuracy for probabilistic predictions, especially in the tails of distributions.
method Introduces a new weighting method for pointwise CRPS learning, optimizing across quantiles.
result Proposed fully adaptive Bernstein online aggregation (BOA) method for pointwise CRPS online learning has optimal convergence properties.

We develop the distance dependent Chinese restaurant process (CRP), a flexible class of distributions over partitions that allows for non-exchangeability. This class can be used to model many kinds of dependencies between data in infinite clustering models, including dependencies across time or space. We examine the pr…

2009-10-06abs ↗pdf ↗

Probabilistic survival predictions from models trained with Maximum Likelihood Estimation (MLE) can have high, and sometimes unacceptably high variance. The field of meteorology, where the paradigm of maximizing sharpness subject to calibration is popular, has addressed this problem by using scoring rules beyond MLE, s…

2018-06-21abs ↗pdf ↗

This paper combines and improves probabilistic forecasts of wind speeds using advanced statistical methods.

problem Improving accuracy and reliability of probabilistic forecasts in wind speed prediction.
method Adapting prediction with expert advice theory to probabilistic forecasts, combining raw or post-processed ensembles, and using CRPS and Jolliffe-Primo tests.
result Combining probabilistic forecasts can lead to more reliable and skillful predictions, as shown by the Jolliffe-Primo test.

Probabilistic forecasts in the form of probability distributions over future events have become popular in several fields of statistical science. The dissimilarity between a probability forecast and an outcome is measured by a loss function (scoring rule). Popular example of scoring rule for continuous outcomes is the …

2019-02-26abs ↗pdf ↗

Develops a forward variable selection method for interpretable random forest models.

problem Interpreting high-dimensional non-parametric models like random forests.
method Forward variable selection using CRPS as loss function, with hypothesis testing at each step.
result Method selects a smaller set of variables that optimizes predictive performance.

Study compares different scoring rules for machine-learned weather forecasts, finding scale-awareness improves forecast realism.

problem Improving the accuracy of machine-learned probabilistic weather forecasts.
method Comparison of scoring rules (CRPS, fair global energy score, graph energy score) and analysis of their impact on forecast field spectra.
result Scale-awareness improves forecast realism, particularly in the tropics.

Paper introduces CRP-O framework for uncertainty quantification in deep operators.

problem Uncertainty quantification in energy-efficient deep learning algorithms, especially in SNNs.
method CRP-O framework using RP networks and SCP, with Gaussian Process Regression for super-resolution.
result Enhanced uncertainty bounds improve UQ estimates compared to existing methods.

A new measure PC allows fair comparison of AIWP and NWP models.

problem Fair comparison of AIWP and NWP model outputs.
method Postprocessing deterministic model outputs with isotonic distributional regression (IDR) and calculating PC as mean CRPS of postprocessed forecasts.
result The GraphCast model outperforms the ECMWF HRES model in WeatherBench 2 data.

Sparse point observations can provide useful local constraints, but their benefit for radar-like fields depends on the training loss, uncertainty representation, and how observation support is encoded in the model.

problem Improving dense radar-field forecasts with sparse point observations
method A multimodal graph neural network nowcasting system over the Nordic radar domain
result Each source improves a different part of the forecast problem

New benchmark predicts cardiometabolic risk from accelerometer data, with varying accuracy.

problem Lack of accurate tabular benchmarks for cardiometabolic risk from accelerometer data.
method Tabular learning methods (ridge regression, XGBoost, TabPFN v2) applied to NHANES data.
result TabPFN v2 achieves best performance, but triglycerides remain largely unpredictable.

Deep forecasting models show output heads significantly improve performance on fat-tailed financial returns.

problem Improving deep learning models for forecasting fat-tailed financial returns.
method Comparison of backbone architectures and output heads (point, Gaussian, Gaussian mixture) on S&P 500 monthly log-returns.
result Switching from point to Gaussian heads improves CRPS by about 1.3 percent, and from Gaussian to mixture adds another 2.4 percent.

Change-point detection (CPD) aims to locate abrupt transitions in the generative model of a sequence of observations. When Bayesian methods are considered, the standard practice is to infer the posterior distribution of the change-point locations. However, for complex models (high-dimensional or heterogeneous), it is n…

2019-10-22abs ↗pdf ↗

CNNs improve wind speed forecasts in the Netherlands.

problem Limited spatial patterns in current post-processing methods.
method Convolutional Neural Networks (CNNs) for spatial wind speed information.
result CNNs produce better probabilistic forecasts with higher Brier skill scores.

Generative model improves noise estimation in stochastic rotating shallow water models.

problem Improving noise estimation in stochastic partial differential equations for fluid dynamics.
method Replaced PCA with a generative model to avoid constraints on stochastic increments.
result Generative model produces better RMSE, CRPS score, and forecast rank histograms.

Generative Bayesian Computation improves surrogates for expensive simulations.

problem Limitations of Gaussian process surrogates in handling complex, non-stationary data.
method Generative Bayesian Computation via Implicit Quantile Networks (IQNs).
result Generative Bayesian Computation outperforms traditional Gaussian process methods across various benchmarks.

U-Cast simplifies AI weather forecasting with a standard U-Net and efficient training.

problem Complex AI models limit accessibility and cost for weather forecasting.
method Simple U-Net backbone, deterministic pre-training, and probabilistic fine-tuning with Monte Carlo Dropout.
result U-Cast matches or exceeds state-of-the-art models in accuracy while reducing training and inference costs.

We are concerned with modeling the strength of links in networks by taking into account how often those links are used. Link usage is a strong indicator of how closely two nodes are related, but existing network models in Bayesian Statistics and Machine Learning are able to predict only wether a link exists at all. As …

2014-02-18abs ↗pdf ↗

Temporal coarse-graining of multi-sector default count data generates effective correlation matrices and rank copulas.

problem Explaining the difference in default dependence between monthly and annual aggregation.
method Dynamic low-rank state-space model with AR(1) latent credit-state factors.
result Effective correlation matrices and rank copulas are generated from monthly default count data.

New approach predicts stock price synchronization using RNNs and LSTMs.

problem Forecasting synchronization of stock prices in the Indian market.
method Utilizing recurrence plots and CRQA for non-linear analysis, RNNs and LSTMs for prediction.
result Accuracy of 0.98 and F1 score of 0.83 in predicting stock price synchronization.

Blade uses diffusion priors to accurately and calibratedly infer complex systems.

problem Derivative-free Bayesian inversion for high-dimensional, nonlinear problems with costly forward models.
method Blade employs an ensemble of interacting particles and diffusion models as priors, querying forward models only through evaluations.
result Blade produces well-calibrated posterior samples that existing methods cannot, improving with more iterations and particles.

New method improves local precipitation predictions using video diffusion.

problem Limited high-resolution local precipitation predictions due to computational costs.
method Extends video diffusion models to capture conditional distribution of high-resolution patterns.
result Method outperforms state-of-the-art baselines in CRPS, MSE, and precipitation distribution.

Deep neural networks forecast financial return distributions accurately.

problem Forecasting probability distributions of financial returns.
method Used 1D CNN and LSTM architectures with custom loss functions to optimize distribution parameters.
result LSTM with skewed Student's t distribution outperformed classical models in multiple evaluation metrics.

Crowdsourcing has been proven to be an effective and efficient tool to annotate large datasets. User annotations are often noisy, so methods to combine the annotations to produce reliable estimates of the ground truth are necessary. We claim that considering the existence of clusters of users in this combination step c…

2014-07-18abs ↗pdf ↗

The study models and forecasts natural gas prices using skewed, heavy-tailed distributions.

problem Modeling and forecasting natural gas prices with heavy tails and conditional heteroscedasticity.
method State-space time series models under skewed, heavy-tailed distributions.
result The proposed model reduces out-of-sample CRPS by 13% for Day-Ahead and 9% for Month-Ahead forecasts.

Tabular foundation models outperform other methods in conditional density estimation across various datasets.

problem Estimating the full conditional distribution of a response given tabular covariates, especially in settings with heteroscedasticity, multimodality, or asymmetric uncertainty.
method Benchmarked three tabular foundation model variants (TabPFN and TabICL) against six CDE baselines on 39 real-world datasets.
result Tabular foundation models achieve the best CDE loss, log-likelihood, and CRPS across all sample sizes, outperforming other methods.