https://www.ml-quant.com/topics/econometrics-forecasting/Forecasting, time series, econometrics and nowcasting. en https://www.ml-quant.com/papers/arxiv/2609.29530/https://www.ml-quant.com/papers/arxiv/2609.29530/Fri, 25 Sep 2026 07:00:00 +0000Proves that training sufficiency, test coverage, and temporal causality cannot be maximized simultaneously in time-series validation, pricing each constraint explicitly. https://www.ml-quant.com/papers/arxiv/2609.25617/https://www.ml-quant.com/papers/arxiv/2609.25617/Fri, 25 Sep 2026 07:00:00 +0000A hierarchical multi-task framework jointly predicts price movement, volatility and volume using liquidity-aware signals, outperforming neural and tree-based baselines on Chinese equity indices. https://www.ml-quant.com/papers/arxiv/2609.23223/https://www.ml-quant.com/papers/arxiv/2609.23223/Fri, 25 Sep 2026 07:00:00 +0000Jointly reconciling hourly price and spread forecasts improves intraday electricity price prediction accuracy by up to 19.7% and battery-arbitrage profits by up to 10.4%. https://www.ml-quant.com/papers/arxiv/2609.26994/https://www.ml-quant.com/papers/arxiv/2609.26994/Fri, 25 Sep 2026 07:00:00 +0000A regime-switching volatility-in-mean VAR reveals that the drivers of growth and inflation tails differ from median dynamics, with macroeconomic uncertainty playing a larger role in downside risk. https://www.ml-quant.com/papers/arxiv/2609.23998/https://www.ml-quant.com/papers/arxiv/2609.23998/Fri, 25 Sep 2026 07:00:00 +0000A stochastic nested fixed-point estimator reduces memory and computational cost for random-coefficients logit demand models, enabling estimation on 100 million markets in hours. https://www.ml-quant.com/papers/arxiv/2609.21587/https://www.ml-quant.com/papers/arxiv/2609.21587/Fri, 25 Sep 2026 07:00:00 +0000Augmenting HAR with a negative moving-average component approximates rough dynamics, outperforming classical models out-of-sample and matching continuous-time rough model accuracy. https://www.ml-quant.com/papers/arxiv/2609.29515/https://www.ml-quant.com/papers/arxiv/2609.29515/Fri, 25 Sep 2026 07:00:00 +0000Introduces a network realized GARCH-Itô model that identifies dynamic volatility transmission among assets using high-frequency data, outperforming recursive forecasts on sector ETFs. https://www.ml-quant.com/papers/ssrn/7486138/https://www.ml-quant.com/papers/ssrn/7486138/Fri, 25 Sep 2026 07:00:00 +0000Using production, employment, and sales data across 426 industries, the research shows that upstream industry signals predict aggregate monthly stock returns with 23.8% out-of-sample R-squared. https://www.ml-quant.com/papers/arxiv/2602.04464/https://www.ml-quant.com/papers/arxiv/2602.04464/Thu, 12 Feb 2026 07:00:00 +0000Including discounted sales of soon-to-expire perishables in demand forecasts leads to underestimating demand, highlighting the need for better forecasting to reduce inventory waste in grocery stores. https://www.ml-quant.com/papers/arxiv/2601.16821/https://www.ml-quant.com/papers/arxiv/2601.16821/Mon, 02 Feb 2026 07:00:00 +0000The article introduces a new Bayesian model that analyzes compositional time series data, effectively handling structural breaks and enhancing forecasting accuracy during these changes. https://www.ml-quant.com/papers/arxiv/2601.10143/https://www.ml-quant.com/papers/arxiv/2601.10143/Fri, 16 Jan 2026 07:00:00 +0000A new adaptive data management system enhances model performance in quantitative finance by constantly updating to reflect market changes, addressing the shortcomings of relying solely on historical data. https://www.ml-quant.com/papers/ssrn/4463148/https://www.ml-quant.com/papers/ssrn/4463148/Sun, 28 Dec 2025 07:00:00 +0000The study promotes the use of Bayesian statistics in finance to better analyze large AI-generated datasets and mitigate misleading significance from traditional methods. https://www.ml-quant.com/papers/arxiv/2511.21873/https://www.ml-quant.com/papers/arxiv/2511.21873/Mon, 01 Dec 2025 07:00:00 +0000A combined A3T-GCN model enhances the accuracy of FTSE100 stock price forecasts by using technical indicators and optimized sequences. https://www.ml-quant.com/papers/arxiv/2504.06550/https://www.ml-quant.com/papers/arxiv/2504.06550/Wed, 12 Nov 2025 07:00:00 +0000U.S. Causal Rhetoric: Growing blame/credit language in congressional tweets changed donation patterns, fueled protests and polarization, and shifted public trust. https://www.ml-quant.com/papers/arxiv/2510.20221/https://www.ml-quant.com/papers/arxiv/2510.20221/Mon, 27 Oct 2025 07:00:00 +0000KG+LLM for Financial Causal Discovery: Combines SEC knowledge graphs, LLM reasoning, and causal discovery to build more accurate finance‑grounded causal models. https://www.ml-quant.com/papers/arxiv/2510.15458/https://www.ml-quant.com/papers/arxiv/2510.15458/Mon, 27 Oct 2025 07:00:00 +0000We show interventionally robust optimization is continuous under a G‑causal Wasserstein distance and introduce a causal normalizing flow that respects this, improving data augmentation for causal prediction and portfolio optimization. https://www.ml-quant.com/papers/repec/wly-jforec-v-44-y-2025-i-2-p-270-280/https://www.ml-quant.com/papers/repec/wly-jforec-v-44-y-2025-i-2-p-270-280/Mon, 27 Oct 2025 07:00:00 +0000Fashion product demand is hard to predict, but machine learning—especially deep learning and ensembles—can make forecasts more accurate. https://www.ml-quant.com/papers/repec/eee-retrec-v-89-y-2021-i-c-s0739885921000068/https://www.ml-quant.com/papers/repec/eee-retrec-v-89-y-2021-i-c-s0739885921000068/Fri, 24 Oct 2025 07:00:00 +0000The study explores new data sources and machine learning techniques to forecast short-term wait times at a US-Mexico border crossing, emphasizing the difficulties of high data variability. https://www.ml-quant.com/papers/arxiv/2509.05922/https://www.ml-quant.com/papers/arxiv/2509.05922/Sat, 13 Sep 2025 07:00:00 +0000The research uses machine learning to identify that the volatility of options-implied risk and market liquidity are key factors causing market lows, challenging simpler models. https://www.ml-quant.com/papers/arxiv/2509.08183/https://www.ml-quant.com/papers/arxiv/2509.08183/Sat, 13 Sep 2025 07:00:00 +0000The paper presents a risk model that combines heavy-tailed priors with chaotic dynamics to predict volatility clustering, fat tails, and extreme events, providing a dual perspective for systemic risk analysis. https://www.ml-quant.com/papers/arxiv/2509.08742/https://www.ml-quant.com/papers/arxiv/2509.08742/Sat, 13 Sep 2025 07:00:00 +0000Multi-modal Forecasting: FinZero, a pre-trained model fine-tuned by the Uncertainty-adjusted Group Relative Policy Optimization method, is introduced in the article, enhancing the accuracy, adaptability, and scalability of financial time series forecasting. https://www.ml-quant.com/papers/arxiv/2508.18921/https://www.ml-quant.com/papers/arxiv/2508.18921/Fri, 29 Aug 2025 07:00:00 +0000The research shows that deep neural networks can accurately forecast financial return distributions and are competitive with traditional models for risk assessment and portfolio management. https://www.ml-quant.com/papers/arxiv/2508.19609/https://www.ml-quant.com/papers/arxiv/2508.19609/Fri, 29 Aug 2025 07:00:00 +0000Time-Series Forecasting Model: FinCast, a new model for financial time-series forecasting, outperforms existing methods by effectively capturing diverse patterns without needing domain-specific adjustments. https://www.ml-quant.com/papers/arxiv/2508.19994/https://www.ml-quant.com/papers/arxiv/2508.19994/Fri, 29 Aug 2025 07:00:00 +0000Wavelet Coherence Architecture: The Coherent Multiplex system uses a multilayer graph to identify and analyze coherence among multiple time series in real-time, with potential uses in neuroscience, finance, and biomedical signal analysis. https://www.ml-quant.com/papers/arxiv/2508.11372/https://www.ml-quant.com/papers/arxiv/2508.11372/Wed, 20 Aug 2025 07:00:00 +0000The research introduces temporal hierarchy forecasting in predicting electricity prices, showing that reconciling forecasts for different time blocks improves accuracy at all levels. https://www.ml-quant.com/papers/arxiv/2508.07192/https://www.ml-quant.com/papers/arxiv/2508.07192/Tue, 12 Aug 2025 07:00:00 +0000The study investigates the eigenvalue of the Wigner random matrix derived from a time series with temporal correlation, discussing the deformation of the semi-circle law and its moments of distribution and convergence. https://www.ml-quant.com/papers/arxiv/2507.15079/https://www.ml-quant.com/papers/arxiv/2507.15079/Fri, 25 Jul 2025 07:00:00 +0000A new method, Isotonic Quantile Regression Averaging (iQRA), for generating probabilistic forecasts from point forecast ensembles in electricity markets, outperforms other methods in reliability and sharpness. https://www.ml-quant.com/papers/arxiv/2507.10588/https://www.ml-quant.com/papers/arxiv/2507.10588/Thu, 17 Jul 2025 07:00:00 +0000A study predicting daily passenger counts for New York City's yellow taxis from 2017-2019 shows a consistent decline in ridership, with the most accurate predictions made using a first-order autoregressive model. https://www.ml-quant.com/papers/arxiv/2507.04074/https://www.ml-quant.com/papers/arxiv/2507.04074/Thu, 10 Jul 2025 07:00:00 +0000The article presents a new Darwinian Agent-Based Modeling method for macroeconomic forecasting, which uses evolutionary principles and simple rules to create realistic economic patterns efficiently. https://www.ml-quant.com/papers/arxiv/2506.22768/https://www.ml-quant.com/papers/arxiv/2506.22768/Thu, 03 Jul 2025 07:00:00 +0000A study found that global residential energy demand rises at temperatures below -5 degrees Celsius and above 30 degrees Celsius, with developed countries more sensitive to high temperatures.