About

Kay Giesecke is a Professor of Management Science & Engineering at Stanford University, where he has been on the faculty since 2005. He is the Founder and Director of Stanford's Advanced Financial Technologies Laboratory, the Director of the Mathematical and Computational Finance Program, and a member of the Institute for Computational and Mathematical Engineering. His research sits at the intersection of technology and finance, focusing on transforming risk intelligence, market oversight, and investment management through the development of stochastic models, statistical machine learning methods, computational algorithms, and software. His key application areas include risk management, market surveillance, fair lending, and sustainable investing, with his work informing financial regulation, guiding institutional practices, and contributing to more transparent, resilient, and equitable financial systems.

Research topics

• Computer Science
• Political Science
• Computer Security
• Economics
• Management science
• Business
• Engineering
• Actuarial science
• Finance
• Data science

Selected publications

• Online Conformal Prediction for Non-Exchangeable Panel Data

  arXiv (Cornell University) · 2026-05-18

  preprintOpen accessSenior author

  Panel data, in which multiple units are repeatedly observed over time, arise throughout science and engineering. Quantifying predictive uncertainty in such settings is challenging because conformal prediction, while distribution-free and model-agnostic, classically relies on exchangeability assumptions that fail under temporal dependence and unit heterogeneity. We propose a simple online conformal framework for non-exchangeable panel data. The method exploits a key feature of online panel prediction: when a forecast is required for one unit, contemporaneous outcomes from related units may already be observed and can serve as a calibration panel. At each round, prediction sets are formed using currently observed calibration units together with two adaptive quantities: history-based similarity weights that emphasize calibration units resembling the target, and an adaptive miscoverage level that is updated whenever target feedback is revealed. This two-state design yields a stepwise coverage bound and a long-run coverage guarantee. Empirically, across synthetic and real panel data sets, the method improves coverage on the worst-covered target units through adaptive interval-width allocation rather than uniform inflation. The two states are complementary: similarity weights protect coverage when target feedback is sparse, while the adaptive level further improves coverage as feedback accumulates.

  PublisherDOI

• Online Conformal Prediction for Non-Exchangeable Panel Data

  ArXiv.org · 2026-05-18

  articleOpen accessSenior author

  Panel data, in which multiple units are repeatedly observed over time, arise throughout science and engineering. Quantifying predictive uncertainty in such settings is challenging because conformal prediction, while distribution-free and model-agnostic, classically relies on exchangeability assumptions that fail under temporal dependence and unit heterogeneity. We propose a simple online conformal framework for non-exchangeable panel data. The method exploits a key feature of online panel prediction: when a forecast is required for one unit, contemporaneous outcomes from related units may already be observed and can serve as a calibration panel. At each round, prediction sets are formed using currently observed calibration units together with two adaptive quantities: history-based similarity weights that emphasize calibration units resembling the target, and an adaptive miscoverage level that is updated whenever target feedback is revealed. This two-state design yields a stepwise coverage bound and a long-run coverage guarantee. Empirically, across synthetic and real panel data sets, the method improves coverage on the worst-covered target units through adaptive interval-width allocation rather than uniform inflation. The two states are complementary: similarity weights protect coverage when target feedback is sparse, while the adaptive level further improves coverage as feedback accumulates.

  PublisherOA PDF

• AICO: Feature Significance Tests for Supervised Learning

  ArXiv.org · 2025-06-29

  preprintOpen access1st authorCorresponding

  Machine learning is central to modern science, industry, and policy, yet its predictive power often comes at the cost of transparency: we rarely know which input features truly drive a model's predictions. Without such understanding, researchers cannot draw reliable conclusions, practitioners cannot ensure fairness or accountability, and policymakers cannot trust or govern model-based decisions. Existing tools for assessing feature influence are limited; most lack statistical guarantees, and many require costly retraining or surrogate modeling, making them impractical for large modern models. We introduce AICO, a broadly applicable framework that turns model interpretability into an efficient statistical exercise. AICO tests whether each feature genuinely improves predictive performance by masking its information and measuring the resulting change. The method provides exact, finite-sample feature p-values and confidence intervals for feature importance through a simple, non-asymptotic hypothesis testing procedure. It requires no retraining, surrogate modeling, or distributional assumptions, making it feasible for large-scale algorithms. In both controlled experiments and real applications, from credit scoring to mortgage-behavior prediction, AICO reliably identifies the variables that drive model behavior, providing a scalable and statistically principled path toward transparent and trustworthy machine learning.

  PublisherOA PDFDOI

• <p>A Set-Sequence Model for Time Series</p>

  SSRN Electronic Journal · 2025-01-01

  preprintOpen accessSenior author
  PublisherDOI

• <p>AICO: Feature Significance Tests for Supervised <span>Learning</span></p>

  SSRN Electronic Journal · 2025-01-01

  preprintOpen access1st authorCorresponding
  PublisherDOI

• Learning Illiquid Asset Prices

  SSRN Electronic Journal · 2025-01-01

  preprintOpen accessSenior author
  PublisherDOI

• Predicting Rating Transitions using Machine Learning

  SSRN Electronic Journal · 2025-01-01

  preprintOpen access
  PublisherDOI

• A Set-Sequence Model for Time Series

  ArXiv.org · 2025-05-16

  preprintOpen accessSenior author

  Many prediction problems across science and engineering, especially in finance and economics, involve large cross-sections of individual time series, where each unit (e.g., a loan, stock, or customer) is driven by unit-level features and latent cross-sectional dynamics. While sequence models have advanced per-unit temporal prediction, capturing cross-sectional effects often still relies on hand-crafted summary features. We propose Set-Sequence, a model that learns cross-sectional structure directly, enhancing expressivity and eliminating manual feature engineering. At each time step, a permutation-invariant Set module summarizes the unit set; a Sequence module then models each unit's dynamics conditioned on both its features and the learned summary. The architecture accommodates unaligned series, supports varying numbers of units at inference, integrates with standard sequence backbones (e.g., Transformers), and scales linearly in cross-sectional size. Across a synthetic contagion task and two large-scale real-world applications, equity portfolio optimization and loan risk prediction, Set-Sequence significantly outperforms strong baselines, delivering higher Sharpe ratios, improved AUCs, and interpretable cross-sectional summaries.

  PublisherOA PDFDOI

• Call for Papers—<i>Management Science</i> Virtual Special Issue on AI for Finance and Business Decisions

  Management Science · 2024-08-20 · 3 citations

  paratext
  PublisherDOI

• Call for Papers—<i>Management Science</i> Virtual Special Issue on Digital Finance

  Management Science · 2024-07-08 · 1 citations

  paratext
  PublisherDOI

Recent grants

• Statistical Methods for Large Interacting Systems and Applications to Mortgage Pools

  NSF · $220k · 2013–2016

Frequent coauthors

• Justin Sirignano

  University of Oxford

  27 shared

• Richard B. Sowers

  University of Illinois Urbana-Champaign

  17 shared

• Konstantinos Spiliopoulos

  16 shared

• Lisa R. Goldberg

  University of California, Berkeley

  15 shared

• Francis A. Longstaff

  14 shared

• Baeho Kim

  Korea University

  13 shared

• Ilya A. Strebulaev

  National Bureau of Economic Research

  12 shared

• Stephen M. Schaefer

  Arizona State University

  12 shared

Education

• Ph.D., Management Science and Engineering

  Stanford University

  2000

• M.S., Management Science and Engineering

  Stanford University

  1995

• B.S., Operations Research and Industrial Engineering

  University of California, Berkeley

  1990

Awards & honors

• JP Morgan AI Faculty Research Award (2019)
• SIAM Financial Mathematics and Engineering Conference Paper…
• Fama/DFA Prize for the Best Asset Pricing Paper in the Journ…
• Gauss Prize of the Society for Actuarial and Financial Mathe…