publications

2024

1. A Note on the Fractional Momentum Strategy

   Saejoon Kim and Hyuksoo Kim

   Applied Economics Letters, 2024

   Abs Bib Website

   The momentum strategy has been existent for a long time and under numerous versions all of which is based on some form of the past return. It was noted recently that this metric of taking the difference of two prices might be a too stringent transformation to achieve stationary. For this matter, a momentum strategy that is not based on simple past return was proposed, called fractional momentum strategy, that retains some memory in the price series. Empirical results have demonstrated significant returns improvement of this strategy over the traditional one albeit under weekly rebalancing. This article investigates the applicability of the strategy under more realistic rebalancing frequencies and finds that the strategy performs inferior to the traditional momentum strategy as rebalancing frequency is decreased.

   @article{saejoonkim2024ael,
     title = {A Note on the Fractional Momentum Strategy},
     journal = {Applied Economics Letters},
     year = {2024},
     doi = {10.1080/13504851.2024.2337329},
     author = {Kim, Saejoon and Kim, Hyuksoo}
   }

2. Estimating Asset Pricing Models in the Presence of Cross-Sectionally Correlated Pricing Errors

   Hyuksoo Kim and Saejoon Kim

   Mathematics, Nov 2024

   Abs Bib Website

   In this study, we propose an adversarial learning approach to the asset pricing model estimation problem which aims to find estimates of factors and loadings that capture time-series covariations while minimizing the worst-case cross-sectional pricing errors. The proposed estimator is defined by a novel min-max optimization problem in which finding a solution is known to be difficult. This contrasts with other related estimators that admit a well-defined analytic solution but do not effectively account for correlations among the pricing errors. To this end, we propose an approximate algorithm based on the alternating optimization procedure and empirically demonstrate that our proposed adversarial estimation framework outperforms other existing factor models, especially when the explanatory power of the pricing model is limited.

   @article{hyuksookim2024mathematics,
     title = {Estimating Asset Pricing Models in the Presence of Cross-Sectionally Correlated Pricing Errors},
     journal = {Mathematics},
     volume = {12},
     number = {21},
     year = {2024},
     month = nov,
     issn = {2227-7390},
     doi = {10.3390/math12213442},
     article-number = {3442},
     author = {Kim, Hyuksoo and Kim, Saejoon}
   }

2022

1. Deep Asset Allocation for Trend Following Investing

   Saejoon Kim and Hyuksoo Kim

   Journal of Experimental & Theoretical Artificial Intelligence, Jul 2022

   Abs Bib Website

   Trend following strategies are well-known to exhibit excellent excess return performance across a wide range of asset classes in various global markets. For the equity asset class in particular, while the securities selection part is relatively a straightforward procedure, the weight allocation part is more debatable and it has traditionally been identified with the equal-weighted allocation strategy. In this paper, we examine security’s own return-based weight allocation strategy for trend following investing and find that this strategy generates superior returns to several well-established weight allocation schemes. In particular, if the true return of the holding period is used ex ante for weight allocation, it is found that this strategy can generate absolutely huge excess returns. Motivated by this finding, we investigate the efficacy of machine learning techniques for regression of securities’ returns to improve the weight calculation in this framework. Empirical results indicate that deep learning provides the means of regression with which largest excess return gains are possible. In particular, it is demonstrated that the return-based weight allocation strategy defined by our proposed deep learning architecture produces substantial abnormal returns outperforming all other broadly recognised weight allocation schemes compared in this paper.

   @article{saejoonkim2022jetai,
     title = {Deep Asset Allocation for Trend Following Investing},
     journal = {Journal of Experimental \& Theoretical Artificial Intelligence},
     volume = {34},
     number = {4},
     pages = {599--619},
     year = {2022},
     month = jul,
     issn = {0952-813X},
     doi = {10.1080/0952813X.2021.1908429},
     author = {Kim, Saejoon and Kim, Hyuksoo}
   }

2. Managing Downside Risk of Low-Risk Anomaly Portfolios

   Hyuksoo Kim and Saejoon Kim

   Finance Research Letters, May 2022

   Abs Bib Website

   In this paper, we present a novel risk-scaling strategy based on a measure of downside risk and investigate its performance on underlying portfolios that are formed on low-risk anomaly. The downside risk-scaling strategy addresses two challenges of the volatility-scaling strategy, namely, underestimation of and indirect management of downside risk. We demonstrate that our downside risk-scaled strategy improves the unscaled underlying low-risk anomaly strategy as well as outperforms volatility-scaled strategy in terms of risk-adjusted return and various performance metrics that are related to downside events.

   @article{hyuksookim2022frl,
     title = {Managing Downside Risk of Low-Risk Anomaly Portfolios},
     journal = {Finance Research Letters},
     volume = {46},
     pages = {102388},
     year = {2022},
     month = may,
     issn = {1544-6123},
     doi = {10.1016/j.frl.2021.102388},
     author = {Kim, Hyuksoo and Kim, Saejoon}
   }

2021

1. Reduction of Estimation Error Impact in the Risk Parity Strategies

   Hyuksoo Kim and Saejoon Kim

   Quantitative Finance, Aug 2021

   Abs Bib Website

   We consider the risk parity strategy in the presence of estimation errors. We show that risk contributions from constituents of this portfolio can be considerably sensitive to estimation errors in the sense that risk contributions are highly uneven on an ex post basis. In particular, we demonstrate that the sensitivity becomes exaggerated if Fama-French factors constitute the portfolio because of their characteristic of having low pairwise correlations. Our work demonstrates that the instability of the out-of-sample risk contributions is associated with a local property with statistical significance near to the constructed portfolio. Based on this observation, we propose a new algorithm for the risk parity strategy to mitigate the sensitivity of the optimized portfolio’s out-of-sample risk contributions from estimation errors. Through empirical study, we find that the portfolio constructed by the proposed algorithm consistently outperforms its competitors in terms of the out-of-sample risk contributions.

   @article{hyuksookim2021qf,
     title = {Reduction of Estimation Error Impact in the Risk Parity Strategies},
     author = {Kim, Hyuksoo and Kim, Saejoon},
     journal = {Quantitative Finance},
     volume = {21},
     number = {8},
     pages = {1351--1364},
     year = {2021},
     month = aug,
     doi = {10.1080/14697688.2021.1881599}
   }

2016

1. A Robot-Assisted Behavioral Intervention System for Children with Autism Spectrum Disorders

   Sang-Seok Yun, Hyuksoo Kim, JongSuk Choi and Sung-Kee Park

   Robotics and Autonomous Systems, Feb 2016

   Abs Bib Website

   The purpose of this paper is to propose and examine the feasibility of a robot-assisted intervention system capable of facilitating social training for children with autism spectrum disorder (ASD) via human–robot interaction (HRI) architecture. Based on the well-known discrete trial teaching (DTT) protocol for the therapy of children with ASD, our control architecture configures four modules–human perception, user input, the interaction manager, and the robot effector–such that the robot system generates differentiated training stimuli using motivation and Stroop paradigms and automatically copes with the child’s response by using reliable human recognition and interaction technologies. Using these configurations, the proposed system performs the role of training the social skills of basic eye contact and reading emotions in children with ASD. By examining reliable performance evaluations and the positive effect of the training process targeting preschoolers with a high functioning level, we then verify that the proposed system can induce a positive transition in the response of children with ASD and the possibility of a labor-saving effect in carrying out autism treatments.

   @article{yun2016ras,
     title = {A Robot-Assisted Behavioral Intervention System for Children with Autism Spectrum Disorders},
     author = {Yun, Sang-Seok and Kim, Hyuksoo and Choi, JongSuk and Park, Sung-Kee},
     journal = {Robotics and Autonomous Systems},
     volume = {76},
     pages = {58--67},
     year = {2016},
     month = feb,
     doi = {10.1016/j.robot.2015.11.004}
   }