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.

169,181 papers · 148 categories

Trend · papers per month

118237355473 · Jun 202019922001200920182026
48 results for reference class forecasting

The paper introduces a method for forecasting corporate sales growth using multiple reference variables.

problem Forecasting corporate sales growth with multiple reference variables.
method Reference class selection using rank-based algorithms and principal components analysis for data dimension reduction.
result Dimension reduced variables with past sales growth rates and operating margins perform well in forecasting.

Hong Kong uses reference class forecasting to improve roadwork project cost and duration estimates.

problem Optimism bias and strategic misrepresentation in infrastructure project forecasts.
method Reference class forecasting applied to 25 roadwork projects in Hong Kong.
result Forecast accuracy distribution and project benchmarking established.

A major source of risk in project management is inaccurate forecasts of project costs, demand, and other impacts. The paper presents a promising new approach to mitigating such risk, based on theories of decision making under uncertainty which won the 2002 Nobel prize in economics. First, the paper documents inaccuracy…

2013-02-14abs ↗pdf ↗

The paper proposes a method to improve sales forecasts by selecting optimal reference classes.

problem Improving forecasts of sales growth exposed to behavioural bias.
method Finding optimal reference classes for each company based on specific predictors and matching forecast distributions to actual sales.
result The past operating margins are strong predictors for future sales distributions.

HierarchicalForecast provides a Python framework for coherent hierarchical forecasting.

problem Ensuring forecasts at disaggregate levels add up to aggregate forecasts.
method Preprocessed datasets, evaluation metrics, and statistical baseline models.
result Python-based reference framework for statistical and ML forecasting.

CDLF predicts product life-cycles in cold-start phases with high accuracy.

problem Forecasting new products in early phases when data is scarce.
method Conditional Diffusion Life-cycle Forecaster (CDLF) combining static descriptors, reference trajectories, and new observations.
result CDLF outperforms classical models in accuracy and probabilistic forecasting.

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

problem Efficiently forecasting Value at Risk (VaR) with long-memory and non-linear realized volatility.
method Loss-based generalized Bayesian inference with Sequential Monte Carlo for model estimation and prediction.
result RNN-HAR model consistently outperforms other VaR forecasting models.

Paper benchmarks and customizes energy forecasting methods.

problem Energy forecasting challenges and differences from traditional time series.
method Collected large-scale load datasets and renewable energy datasets. Developed feature engineering and customized loss functions.
result Comprehensive evaluation of 21 forecasting methods in energy datasets.

Classifies load forecasting studies by forecasting problem.

problem Selecting appropriate load forecasting techniques and methodologies.
method Classification based on two forecasting problem parameters.
result Synthetic view of relevant forecasting techniques and methodologies.

Paper uses ML for electricity price forecasting using order book data.

problem Forecasting German electricity spot market prices.
method Developed feature extraction for order book data, used cross-validation, compared neural networks and random forests to statistical models.
result Machine learning models outperform traditional approaches.

New models analyze how ECB's unconventional policies affect stock market volatility.

problem Analyzing the impact of ECB's unconventional policies on stock market volatility.
method Developed MEM with Asymmetry and Policy effects (MAP) models to separate base volatility from policy effects.
result Significant improvement in forecasting power after Expanded Asset Purchase Programme implementation.

This paper reviews forecast combinations over 50 years, highlighting their evolution and utility.

problem Improving forecast accuracy through combining multiple forecasts.
method Evolution of forecast combination methods, from simple to sophisticated.
result Forecast combinations have become a mainstream approach in forecasting.

This paper improves forecast stability without sacrificing accuracy using dynamic loss weighting.

problem Rolling origin forecast instability in time series forecasting.
method Dynamic loss weighting algorithms applied to the N-BEATS model.
result Dynamic loss weighting can further improve forecast stability without compromising accuracy.

Deep learning predicts tropical cyclone tracks efficiently.

problem Forecasting tropical cyclone trajectories with high precision and speed.
method Fused neural network model using past trajectory data and reanalysis atmospheric images.
result Deep learning can provide valuable and complementary predictions for tropical cyclone tracks.

Simplifies RF predictions by focusing on a subset of nearest neighbors.

problem Improving interpretability and performance of RF-based forecast distributions.
method Sparsifying RF-based forecast distributions by focusing on a small subset of nearest neighbors.
result Simplified RF predictions can be similar to or exceed original ones in forecasting performance.

A monitoring procedure improves machine learning forecasts for digital platforms.

problem Maintaining accurate and stable forecasts for data streams at digital platforms.
method Developed a monitoring procedure to determine when to retrain machine learning algorithms.
result Monitor-based retraining produces accurate forecasts compared to benchmarks.

Paper introduces SMM for forecasting multiple time series with missing values.

problem Forecasting multiple time series with missing and noisy values.
method Sliding Mask Method (SMM) using Non-negative Matrix Factorization (NMF).
result The method outperforms state-of-the-art methods in time series forecasting.

A simple baseline outperforms deep learning methods in transportation forecasting.

problem The importance of stationarity and recurrent patterns in transportation data.
method A naive baseline based on average weekly patterns and linear regression.
result The baseline method achieves comparable or better results than state-of-the-art deep learning approaches.

A new deep learning method improves weather forecasting accuracy and uncertainty quantification.

problem Improper initial states in NWP lead to unsatisfactory weather forecasting performance.
method Data-driven approach with negative log-likelihood error (NLE) loss function and deep ensemble strategies.
result The proposed method significantly improves accuracy by 47.76% compared to NWP.

A graph neural network improves multivariate post-processing of ensemble forecasts.

problem Systematic biases in ensemble forecasts and loss of dependencies across forecast dimensions.
method A composite-Loss Graph Neural Network (dualGNN) trained with a composite loss function combining ES and VS.
result The dualGNN outperforms traditional methods in multivariate verification metrics and captures spatial relationships.

A framework for forecasting high-dimensional time-series data using clustering.

problem Forecasting high-dimensional time-series data with intra-cluster similarity.
method Three-stage framework: univariate time series parameter estimation, clustering, multivariate time series parameter computation.
result Framework achieves state-of-the-art results on benchmark datasets, sometimes outperforming deep-learning-based approaches.

This paper develops a method to select a reference contract for multi-contract quoting to minimize execution risk.

problem Minimizing execution risk in multi-contract quoting sequences.
method Develops a diagnostic framework using order-flow Hawkes forecasts and CLF to select a stable reference contract.
result Event-history and LOB-state signals offer complementary views for reference-contract selection.

X-Trend quickly adapts to new financial regimes, increasing Sharpe ratio by 18.9%.

problem Adapting to rapidly changing financial market conditions.
method Few-shot learning and cross-attention mechanism.
result X-Trend increases Sharpe ratio by 18.9% over a neural forecaster and 10-fold over a conventional strategy.

Deep ESN models forecast spatio-temporal data with uncertainty quantification.

problem Complex nonlinear dynamics in spatio-temporal systems are hard to model.
method Deep ensemble ESN models using bootstrap and hierarchical Bayesian frameworks.
result Models produce forecasts and uncertainty measures for spatio-temporal data.

System detects relevant financial news and predictions from unstructured text.

problem Manual extraction of relevant financial information from news is cumbersome and error-prone.
method Topic modeling with LDA, co-reference resolution, multi-paragraph segmentation, and temporal analysis.
result ROUGE-L values for relevant text and predictions/forecasts were 0.662 and 0.982, respectively.

Bob predicts a future observation based on a sample of size one. Alice can draw a sample of any size before issuing her prediction. How much better can she do than Bob? Perhaps surprisingly, under a large class of loss functions, which we refer to as the Cover-Hart family, the best Alice can do is to halve Bob's risk. …

2012-06-15abs ↗pdf ↗

Bayesian LSTM model improves VaR and ES forecasting accuracy.

problem Joint forecasting of Value at Risk (VaR) and Expected Shortfall (ES).
method Hybrid model combining LSTM for time series dynamics and Asymmetric Laplace quasi-likelihood for joint likelihood.
result The LSTM-AL model outperforms existing models in VaR and ES forecasting accuracy.

FreDN separates trends and periodicities in non-stationary time series forecasts.

problem Spectral entanglement and computational burden in frequency-domain methods for non-stationary time series.
method FreDN introduces a learnable Frequency Disentangler module to separate trend and periodic components directly in the frequency domain, and uses a ReIm Block to reduce complexity.
result FreDN outperforms state-of-the-art methods by up to 10% on long-term forecasting benchmarks.