Columbia-NYU Financial Engineering Colloquium: Graeme Baker and Kay Giesecke

Graeme Baker

Columbia University

Title

Joint Calibration of Credit Default Swaps and Index Tranches via Elastically Stopped Lévy Processes

Abstract

We model a firm's default as the first time the running supremum of a latent Lévy distress process crosses an independent exponential barrier. This elastic stopping mechanism exhibits the interpretable latent state of a structural model, as well as the totally inaccessible default times of an intensity model. For spectrally positive drivers with phase-type jumps, single-name CDS curves are priced by an explicit Laplace transform, and a common jump factor yields a multi-credit model with simultaneous defaults that is simulated exactly. On daily CDX panels, the model outperforms structural, affine, and copula benchmarks on both constituents and index tranches. We then add mean-field contagion by coupling firms through the default fraction. Without the common jump factor, the large-population limit is deterministic, and the tranche dependence is lost; with it, the limit retains the dependence observed at finite population size. Based on joint work with Agostino Capponi: https://arxiv.org/abs/2608.10321

Bio

Graeme Baker is a Term Assistant Professor in the Department of Statistics at Columbia University, where he also teaches in the M.A. in Mathematical Finance program. He received his PhD in Applied and Computational Mathematics from Princeton University, advised by Mykhaylo Shkolnikov. His research concerns particle systems with singular interactions, probabilistic solutions of free boundary problems, and the calibration of stochastic models to credit and equity market data.

Kay Giesecke

Hasso Plattner Institute

Title

Learning Illiquid Asset Prices

Abtract

Many asset classes—including private equity, real estate, corporate, municipal, and mortgage bonds, as well as structured products—are illiquid, with prices that are observed only sporadically. Yet frequent, objective, and accurate price estimates are essential for portfolio valuation, investment decision-making, and regulatory compliance. This paper develops point and interval estimators for illiquid asset prices, offering both consistency guarantees and non-asymptotic coverage for the resulting intervals. The estimators are based on a scalable, assumption-light semiparametric conditional factor model for the quantiles of asset prices. Empirical results for mortgage-backed securities demonstrate the method’s effectiveness and represent the first treatment of illiquid MBS pricing in the literature, highlighting the bond and market characteristics that drive prices. This is joint work with Junting Duan (HKU) and Yang Fan (Cubist).

Bio

Kay Giesecke is Professor of Artificial Intelligence & Quantitative Finance at Hasso Plattner Institute. His research lies at the intersection of AI, computation, and finance. He is developing a new research initiative focused on engineering intelligent financial systems that can learn, reason, discover, decide, and act. Before joining HPI, he spent more than two decades on the faculty of Stanford University, where he founded and led the Advanced Financial Technologies Laboratory and directed the Mathematical and Computational Finance Program. In 2020, he founded Infima Technologies, a venture-backed AI company serving fixed-income markets, and later led the company through its acquisition in 2024. He has published across operations research, machine learning, econometrics, and financial economics and has received several distinctions, including the Fama/DFA Prize and the JPMorgan AI Faculty Research Award. A former Editor of Management Science (Finance Area), he currently serves as Associate Editor for Mathematical Finance, the SIAM Journal on Financial Mathematics, and several other journals.