Novel framework uses causality for financial forecasting.
problem Balancing invariance and prediction accuracy in financial time series.
method Causality-inspired models for forecasting asset returns.
result Efficacy in stable and accurate predictions, especially in turbulent markets.
CauSTream forecasts streamflow by integrating causal graphs for better interpretability.
problem Streamflow forecasting lacks interpretability and generalization due to fixed causal models.
method CauSTream learns causal graphs for meteorological forcings and routing dependencies.
result CauSTream outperforms existing methods, especially at longer forecast windows.
Study on forecasting methods and their causal implications.
problem Understanding the difference between statistical and causal risks in forecasting models.
method Introduce causal learning theory for forecasting, obtain uniform convergence bounds for VAR models.
result First theoretical guarantees for causal generalization in time-series forecasting.
DoFlow models time series data for causal forecasting and anomaly detection.
problem Forecasting and causal reasoning in multivariate time series.
method Flow-based generative model over causal DAGs.
result Accurate interventional and counterfactual forecasting, anomaly detection.
In many scientific fields, such as economics and neuroscience, we are often faced with nonstationary time series, and concerned with both finding causal relations and forecasting the values of variables of interest, both of which are particularly challenging in such nonstationary environments. In this paper, we study c…
Paper proposes a new method for demand forecasting in pricing contexts.
problem Demand forecasting in pricing contexts, especially in a profit optimal manner.
method Combines Double Machine Learning for causal inference and transformer-based forecasting models.
result Our method outperforms other forecasting methods in off-policy settings.
Paper proposes forecast-necessity testing for accurate causal interpretation in nonlinear time-series models.
problem Misinterpretation of causal scores from nonlinear models as regression coefficients.
method Systematic edge ablation and forecast comparison to evaluate causal necessity.
result Causal relationships with similar scores can differ in their necessity for accurate prediction.
GCRL learns causal factors for motion forecasting, improving out-of-distribution prediction.
problem Sensitivity to out-of-distribution data in conventional supervised learning methods.
method Generative Causal Representation Learning (GCRL) leveraging causality for knowledge transfer.
result Significantly outperforms prior models on out-of-distribution prediction.
Improved causal inference with panel data using deep learning.
problem Causal inference challenges in social science with panel data.
method Adapted N-BEATS deep neural architecture for time series forecasting.
result SyNBEATS estimator outperforms existing methods in panel data settings.
Surveying machine learning methods for economic forecasting.
problem Improving accuracy of economic forecasts using machine learning.
method Nowcasting, textual data, panel and tensor data, high-dimensional Granger causality tests, time series cross-validation, classification with economic losses.
result Recent advances in machine learning methods enhance economic forecasting accuracy.
Study causal financial signals for non-stationary markets, improving short-term forecasts.
problem Short-term forecasting in non-stationary financial markets under causal constraints.
method Construct causal signals from heterogeneous micro-features using causal centering, linear aggregation, Kalman filter, and forward-like operator.
result Causally constructed observables can exhibit substantial economic relevance in specific regimes but degrade under regime shifts.
DeepPPMNet forecasts EMS demand and performs causal analyses for policy-making.
problem Accurate prediction and causal analysis of EMS demand for effective policy-making.
method DeepPPMNet, a LSTM-based framework, globally forecasts and analyzes causal relationships using Granger causality.
result DeepPPMNet outperforms traditional methods in forecasting EMS demand and policy-making.
LAVARNET predicts multivariate time series by estimating causal variable relationships.
problem Forecasting multivariate time series requires understanding causal interrelationships among variables.
method LAVARNET is a neural network architecture that estimates causal effects and predicts future values.
result LAVARNET outperforms other models on various real-world data sets.
Timer-XL predicts multidimensional time series using a unified Transformer approach.
problem Unified time series forecasting across various tasks and contexts.
method Decoder-only Transformers with a universal TimeAttention mechanism and deft position embedding.
result State-of-the-art performance across multiple forecasting benchmarks.
New method uses neural networks to forecast spatial-temporal data.
problem Probabilistic forecasting of spatio-temporal data with causal structure.
method MMAF-guided learning with ensemble of stochastic feed-forward neural networks.
result Forecasting remains calibrated across multiple time horizons.
TCFimt forecasts causal effects of multiple interventions from individual data.
problem Estimating causal effects of temporal multi-interventions from individual data.
method TCFimt uses adversarial tasks in seq2seq framework to alleviate bias and contrastive learning to decouple effects.
result TCFimt outperforms state-of-the-art methods in predicting future outcomes and choosing optimal treatments.
New method uses MMAF-guided learning for spatio-temporal probabilistic forecasts.
problem Probabilistic forecasting of spatio-temporal data with causal structure.
method Generalized Bayesian methodology, MMAF-guided learning, ensemble of stochastic feed-forward neural networks.
result Forecast performance comparable to, and sometimes better than, deep learning architectures.
Transformer-based method for causal discovery with prior knowledge integration.
problem Complex nonlinear dependencies and spurious correlations in time series data.
method Multi-layer Transformer forecaster with gradient-based causal structure extraction and attention masking for prior knowledge integration.
result Significant improvement in causal discovery and causal lag estimation compared to state-of-the-art methods.
Improved Granger causality method for dynamic time series data.
problem Traditional Granger causality method assumes constant causalities, failing to model dynamic causalities.
method Dynamic window-level Granger causality (DWGC) method with causality indexing.
result Improved DWGC method better detects window-level causalities.
CSHT predicts financial returns from news using a novel transformer model on a sphere.
problem Financial forecasting from news and sentiment.
method Granger-causal hypergraph structure, Riemannian geometry, causally masked Transformer attention.
result CSHT outperforms baselines in return prediction, regime classification, and asset ranking.
This paper rates robustness of multi-modal time-series forecasting models.
problem Robustness of AI systems in time-series forecasting is crucial for stakeholders.
method Causal analysis to assess robustness of MM-TSFM models.
result Multi-modal forecasting models are more robust than numeric models.
Incorporating nonlinearity is paramount to predicting the future states of a dynamical system, its response to shocks, and its underlying causal network. However, most existing methods for causality detection and impulse response, such as Vector Autoregression (VAR), assume linearity and are thus unable to capture the …
Introduces recency bias to improve time-series forecasting.
problem Lack of recency bias in standard Transformer attention for time-series data.
method Reweights attention scores with a smooth heavy-tailed decay to emphasize nearby observations.
result Recency-biased attention consistently improves sequential modeling and achieves competitive performance on time-series forecasting benchmarks.
This paper reviews causal inference methods for time series data.
problem Estimating treatment effects and identifying causal relations from time series data.
method Comprehensive review of approaches for treatment effect estimation and causal discovery.
result Provides a list of evaluation metrics and datasets for time series causal inference.
DeepVol uses high-frequency data to forecast volatility, outperforming traditional methods.
problem Improving volatility forecasting using high-frequency data.
method Dilated Causal Convolutions applied to high-frequency financial time-series.
result DeepVol outperforms traditional methods in forecasting day-ahead volatility.
CRC improves multivariate forecasting accuracy without risking performance degradation.
problem Systematic errors and lack of guarantees in multivariate forecasters.
method CRC uses a causality-inspired encoder and hybrid corrector with a safety mechanism.
result CRC consistently improves accuracy and ensures high non-degradation rates.
Bayesian method tests Granger causality in functional time series.
problem Testing Granger causality between functional time series.
method Bayesian dynamic linear models (DLM) and Bayes Factor.
result Captures Granger causality between yield curves and weather conditions.
Develops a new method to discover causal relationships from nonstationary time series data.
problem Challenges in inferring causal relationships from observational data, especially for nonstationary time series.
method State-Dependent Causal Inference (SDCI) for conditionally stationary time series.
result SDCI can recover underlying causal dependencies with provable identifiability for state-dependent causal structures.
We introduce the Pricing Engine package to enable the use of Double ML estimation techniques in general panel data settings. Customization allows the user to specify first-stage models, first-stage featurization, second stage treatment selection and second stage causal-modeling. We also introduce a DynamicDML class tha…
This paper forecasts cryptocurrency log-returns using LASSO-VAR and sentiment analysis.
problem Forecasting log-returns of cryptocurrencies using social media sentiment.
method LASSO-VAR model combined with Twitter and Reddit sentiment data.
result The model predicts the correct direction of cryptocurrency returns more than 50% of the time.
CAST predicts distribution-valued time series by stabilizing and transporting simplex-supported successors.
problem Forecasting distribution-valued time series with structural failure modes.
method CAST (Causal Anchored Simplex Transport) uses successors retrieved from causal context, stabilized with a persistence anchor, and locally transported on ordered supports.
result CAST outperforms baselines on eleven public and simulated benchmarks, achieving best average rank on both one-step KL and autoregressive rollout JSD.
DBNs improve VaR forecasting compared to traditional models, but SVaR forecasts are conservative.
problem Forecasting VaR and SVaR using dynamic Bayesian networks.
method DBN framework applied to S&P 500 index returns, comparing to autoregressive models and historical simulation.
result DBNs achieve comparable VaR forecasting accuracy to historical simulation models, but SVaR forecasts remain conservative.
Graph neural networks improve volatility forecasts and portfolio performance.
problem Improving volatility forecasting for better portfolio performance.
method Compared Heterogeneous Autoregressive and Long Short-Term Memory models with GraphSAGE models built on rolling correlation, sector, and Granger-causal graphs.
result GraphSAGE models with macro regime features outperform other models in terms of forecast accuracy, ranking quality, and portfolio Sharpe ratio.
Graph Neural Network improves causal inference in dynamic systems.
problem Identifying causal relations among multi-variate time series.
method Graph Neural Network approach with score-based method.
result Graph Neural Network significantly outperformed other methods in dynamic Bayesian network inference.
MetaPhysiCa tackles robust physics-informed machine learning for OOD tasks.
problem Designing robust PIML methods for OOD forecasting tasks in physics.
method Meta-learning procedure for causal structure discovery including invariant risk minimization.
result Significantly outperforms existing PIML and deep learning methods in OOD tasks.
Proposes DCNAR for dynamic causal inference from neural time series.
problem Uncertainty and evolution of causal structure in real-world domains.
method Two-stage neural causal modeling integrating discovery and inference.
result Dynamic causal inferences are more stable and meaningful than alternatives.
We present a probabilistic forecasting framework based on convolutional neural network for multiple related time series forecasting. The framework can be applied to estimate probability density under both parametric and non-parametric settings. More specifically, stacked residual blocks based on dilated causal convolut…
Unified model forecasts epidemics with spatial and temporal dynamics.
problem Limited accuracy in traditional models and lack of interpretability in deep learning models.
method CSTGNN integrates Spatio-Contact SIR model with Graph Neural Networks.
result Effective spatiotemporal epidemic forecasting with interpretability.
We present a new method for forecasting systems of multiple interrelated time series. The method learns the forecast models together with discovering leading indicators from within the system that serve as good predictors improving the forecast accuracy and a cluster structure of the predictive tasks around these. The …
New framework IDOL identifies latent causal processes with instantaneous relations from time series data.
problem Identifying latent causal processes with instantaneous relations from time series data.
method Sparse influence constraint and variational inference architecture with sparsity regularization.
result Our method can identify latent causal processes with instantaneous relations.
Proposes a deep learning model for probabilistic forecasting that is also interpretable.
problem Inability to explain predictions of neural network-based time series forecasting methods.
method Deep Autoregressive Networks (DANLIP) for locally interpretable probabilistic forecasting.
result DANLIP provides interpretable predictions with comparable performance to state-of-the-art methods.
Method identifies causal interactions between time series using extreme eigenvalue variability.
problem Detecting causal interactions between time series.
method Largest eigenvalue of lagged correlation matrices, measuring causal interactions through variability.
result The method outperforms traditional Granger causality tests in detecting structural changes.
FOCUS method forecasts counterfactuals in panel data with time series dynamics.
problem Forecasting unobserved potential outcomes in causal inference with missing entries and latent factors.
method FOCUS extends matrix completion methods by leveraging time series dynamics of latent factors.
result FOCUS method outperforms existing benchmarks in predicting future counterfactuals.
New method identifies cause-effect relations in multivariate time series data.
problem Identifying cause-effect relations in multivariate time series data.
method Fictitious vector autoregressive model to identify long-run relations and causality strength.
result High accuracy in identifying true cause-effect relations in simulations and climate change analysis.
Paper introduces TS-GPT for engineering time series forecasting.
problem Engineering time series require causal operations, unlike linguistic data.
method Innovations representation theory, Generative Pre-trained Transformer.
result TS-GPT effectively forecasts real-time locational marginal prices.
Causality graphs are routinely estimated in social sciences, natural sciences, and engineering due to their capacity to efficiently represent the spatiotemporal structure of multivariate data sets in a format amenable for human interpretation, forecasting, and anomaly detection. A popular approach to mathematically for…
Proposes PEID for analyzing synergistic causation in complex systems.
problem Challenges in identifying and analyzing synergistic causation in complex systems.
method Partial Effective Information Decomposition (PEID) framework.
result Unified and computable characterization of synergistic causal relations.
This paper uses CausalGANs and RL with LLM to predict bond yields.
problem Challenges in financial bond yield forecasting due to data scarcity and market conditions.
method Proposes a novel framework combining CausalGANs, RL, and LLM for synthetic data generation and trading signals.
result Improves forecasting performance over existing methods with low Mean Absolute Error.