Financial Time Series

Ask about this topic

Overview

Financial time series research analyzes sequential market and economic measurements—such as prices, returns, trading activity, yields, and volatility—to understand dynamics and support forecasting, risk management, and portfolio decisions. Approaches range from econometric baselines such as random walks, ARIMA, and GARCH to boosting, recurrent networks, Transformers, and state-space models; other methods use regime-aware experts, cross-asset relationships, news or social sentiment, and generative models for synthetic data. Researchers use chronological walk-forward tests and task-specific measures, including forecast error, calibration, tail risk, transaction costs, and risk-adjusted returns, rather than relying only on generic accuracy. A central tension is that markets are noisy and non-stationary: complex models may capture regimes and extra signals, yet often fail to beat simple baselines or produce profitable decisions, leaving robustness and out-of-sample economic value as open challenges.

Latest papers

CardsList

95 papers13 papers in the last four weeks, up 63% on the four weeks before. 0.1% of all new papers.

  1. How do standard time series forecasting benchmarks translate to finance? A broad study of risk adjusted performance

    Mar 2, 2026Adir Saly-Kaufmann, Kieran Wood, Jan-Peter Calliess +1Financial Time SeriesTime Series Forecasting

    Machine Learning Research Group, Department of Engineering Science, University of Oxford · Oxford-Man Institute of Quantitative Finance, University of Oxford

  2. Financial time series augmentation using transformer based GAN architecture

    Feb 19, 2026Andrzej Podobiński, Jarosław A. ChudziakTime Series Data AugmentationFinancial Time Series

    Faculty of Electronics and Information Technology, Warsaw University of Technology, Warsaw, Poland

  3. Cross-Sectional Asset Retrieval via Future-Aligned Soft Contrastive Learning

    Feb 11, 2026Hyeongmin Lee, Chanyeol Choi, Jihoon Kwon +8Contrastive LearningInformation Retrieval

    Seoul National University of Science and Technology Seoul, Republic of Korea · LinqAlpha Cambridge, MA, United States · Massachusetts Institute of Technology Cambridge, MA, United States +7

  4. KAN vs LSTM Performance in Time Series Forecasting

    Nov 23, 2025Tabish Ali Rather, S M Mahmudul Hasan Joy, Nadezda Sukhorukova +1Time Series ForecastingNon-Stationary Time Series Forecasting

    Department of Mathematics, Swinburne University of Technology, Hawthorn, 3122, Victoria, Australia.

  5. Realised Volatility Forecasting: Machine Learning via Financial Word Embedding

    Aug 1, 2021Eghbal Rahimikia, Stefan Zohren, Ser-Huang PoonFinancial ForecastingTime Series Forecasting

    Alliance Manchester Business School at the University of Manchester · Department of Engineering Science, University of Oxford · Alliance Manchester Business School, University of Manchester