arXiv research

Paper compares two forecasters using novel online inference methods.

A new framework detects forecast model inadequacies using online monitoring of forecast errors.

Paper analyzes Nyström regularization for time series forecasting with sequential sub-sampling.

Survey of algorithms to correct past mistakes in prediction.

Introduces recency bias to improve time-series forecasting.

New method recalibrates VaR for option books, reducing forecast errors.

SOR-Mamba improves Mamba for robust time series forecasting by minimizing channel order bias.

Improved LSTM cell for high-frequency trading forecasts.

A new CNN approach for time series forecasting outperforms traditional RNNs.

RNN-HAR model improves VaR forecasting with long-memory and non-linear dynamics.

Time series forecasting is difficult. It is difficult even for recurrent neural networks with their inherent ability to learn sequentiality. This article presents a recurrent neural network based time series forecasting framework covering feature engineering, feature importances, point and interval predictions, and for…

Unified deep sequential and state-space models for robust option pricing with uncertainty.

The purpose of this paper is to propose a time-varying vector autoregressive model (TV-VAR) for forecasting multivariate time series. The model is casted into a state-space form that allows flexible description and analysis. The volatility covariance matrix of the time series is modelled via inverted Wishart and singul…

Proposes handling ambiguity in sequential data predictions.

Combines deep state space models with diffusion models for better forecasting and capturing latent dynamics

Prequential posteriors tackle data assimilation for deep generative forecasting models.

Structured subsampling improves privacy in deep time series forecasting.

Develops BPDS for better financial portfolio decisions.

Generative networks minimize predictive scoring rules for probabilistic forecasting.

Parity calibration aims to predict increase-decrease events, not values.

New framework optimizes forecasting and decision-making in dynamic systems.

We develop a new DTSM with nonlinearities using Gaussian Processes for better interest rate forecasting.

In this paper we develop a Bayesian procedure for estimating multivariate stochastic volatility (MSV) using state space models. A multiplicative model based on inverted Wishart and multivariate singular beta distributions is proposed for the evolution of the volatility, and a flexible sequential volatility updating is …

Production forecasting is a key step to design the future development of a reservoir. A classical way to generate such forecasts consists in simulating future production for numerical models representative of the reservoir. However, identifying such models can be very challenging as they need to be constrained to all a…

Dynamic probabilistic forecasts guide optimal decisions in uncertain processes.

Novel time series forecasting method using sliding window signatures.

Paper introduces COBRA variations for multivariate time series forecasting.

This work addresses identifiability in sequential data with switching dynamics, introducing a new estimator.

STOIC improves energy demand forecasting with reliable uncertainty estimates.

Post-training corrections boost time-series forecasting accuracy.

The book explores universal time-series forecasting using mixture predictors.

New RNN model forecasts unseen time series with little training data.

This work combines recurrent models with diffusion for probabilistic time series forecasting.

New method discovers accurate time series models using SMC and MCMC.

Generative adversarial network for probabilistic forecasting of random systems.

Combines machine learning and data assimilation for improved forecasting.

A new model integrates LSTM and copulas for high-dimensional financial data.

We propose a framework for general probabilistic multi-step time series regression. Specifically, we exploit the expressiveness and temporal nature of Sequence-to-Sequence Neural Networks (e.g. recurrent and convolutional structures), the nonparametric nature of Quantile Regression and the efficiency of Direct Multi-Ho…

New calibration measure SSCE ensures truthful prediction, unlike existing measures.

ResCP uses reservoir computing to create efficient, scalable time series prediction intervals.

The paper studies the distance from calibration in sequential prediction, proving upper and lower bounds.

We establish optimal rates for online regression for arbitrary classes of regression functions in terms of the sequential entropy introduced in (Rakhlin, Sridharan, Tewari, 2010). The optimal rates are shown to exhibit a phase transition analogous to the i.i.d./statistical learning case, studied in (Rakhlin, Sridharan,…

Multivariate boosted trees improve forecasting and control by capturing correlated predictions.

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

Paper presents a method for probabilistic load forecasting using adaptive online learning.

ADD-THIN improves TPP forecasting by handling long-term data sequences.

This paper establishes minimax rates for online regression with arbitrary classes of functions and general losses. We show that below a certain threshold for the complexity of the function class, the minimax rates depend on both the curvature of the loss function and the sequential complexities of the class. Above this…

We consider the setting of sequential prediction of arbitrary sequences based on specialized experts. We first provide a review of the relevant literature and present two theoretical contributions: a general analysis of the specialist aggregation rule of Freund et al. (1997) and an adaptation of fixed-share rules of He…