Jeong-Kyu Heo
Sungkyunkwan University · Economics, Econometrics and Finance
About the Lab
Professor Jeong-Kyu Heo's research lab specializes in quantitative finance and financial engineering, focusing on the development of advanced computational methods for pricing complex derivatives and solving high-dimensional stochastic control problems in continuous time. The lab integrates modern deep learning techniques—particularly recurrent networks and neural network-based policy optimization—with classical stochastic control theory, such as Pontryagin’s Maximum Principle, to enhance accuracy and efficiency in option pricing and portfolio optimization. Key research directions include the application of neural networks to solve Volterra and backward SDEs, the design of stable and interpretable algorithms for American options, and the construction of multiscale models for volatility surface modeling.
Research Overview
Research Output Trend
Figures are computed from collected data and may differ slightly.
Selected Papers
15The number of tailor-made hybrid structured products has risen more prominently to fit each investor’s preferences and requirements as they become more diversified. The structured products entail synthetic derivatives such as combinations of bonds and/or stocks conditional on how they are backed up by underlying securities, stochastic volatility, stochastic interest rates or exchanges rates. The complexity of these multi-asset structures yields lots of difficulties of pricing the products. Becau
Efficiently determining a price and optimal exercise boundary for an American option is a critical subject in the financial sector. This study introduces a novel application of long short-term memory neural networks to solve a relevant Volterra equation, enhancing the accuracy and efficiency of American option pricing. The proposed approach outperforms traditional numerical techniques, including finite difference methods, binomial trees, and Monte Carlo methods, delivering an impressive speed im
We present Pontryagin-Guided Direct Policy Optimization (PG-DPO), a framework for solving continuous-time portfolio optimization problems involving both consumption and investment decisions. Integrating Pontryagin's Maximum Principle (PMP) within a neural network pipeline, PG-DPO bypasses traditional value function approximation and directly optimizes policy parameters using adjoint processes associated with the current policy, computed via automatic differentiation. An optional alignment penalt
Under the Generalized Extreme Value (GEV) model, Markose and Alerton (2011) derived the analytic form solutions for vanilla options, and also removed the distortion of the market only with an additional parameter. In this paper, we use the technique in Rubinstein and Reiner (1991) to get the analytic form solutions for barrier options by introducing the Corrected BS (CBS) modelthe BS model close to the GEV model. By introducing CBS volatility we show that barrier option prices are continuous wit
We introduce the Pontryagin-Guided Direct Policy Optimization (PG-DPO) framework for high-dimensional continuous-time portfolio choice. Our approach combines Pontryagin's Maximum Principle (PMP) with backpropagation through time (BPTT) to directly inform neural network-based policy learning, enabling accurate recovery of both myopic and intertemporal hedging demands--an aspect often missed by existing methods. Building on this, we develop the Projected PG-DPO (P-PGDPO) variant, which achieves ne
Research Areas
Dive deeper into Jeong-Kyu Heo's research on Nubint
Open this lab's papers in the app to read with AI, summarize, and cite in your writing.