Woo Chang Kim

• Korea Advanced Institute of Science and Technology

In decision-making problems under uncertainty, predicting unknown parameters is often considered independent of the optimization part. Decision-focused learning (DFL) is a task-oriented framework that integrates prediction and optimization by adapting the predictive model to give better decisions for the corresponding task. Here, an inevitable chal...

In practice, including large number of assets in mean-variance portfolios can lead to higher transaction costs and management fees. To address this, one common approach is to select a smaller subset of assets from the larger pool, constructing more efficient portfolios. As a solution, we propose a new asset selection heuristic which generates a pre...

Two-stage stochastic programs (2SPs) are important tools for making decisions under uncertainty. Decision-makers use contextual information to generate a set of scenarios to represent the true conditional distribution. However, the number of scenarios required is a barrier to implementing 2SPs, motivating the problem of generating a small set of su...

Mean–variance optimization, introduced by Markowitz, is a foundational theory and methodology in finance and optimization, significantly influencing investment management practices. This study enhances mean–variance optimization by integrating machine learning-based anomaly detection, specifically using GANs (generative adversarial networks), to id...

Our study harnesses the power of natural language processing (NLP) to explore the relationship between dietary patterns and metabolic health outcomes among Korean adults using data from the Seventh Korea National Health and Nutrition Examination Survey (KNHANES VII). Using Latent Dirichlet Allocation (LDA) analysis, we identified three distinct die...

This study presents a novel approach to predicting price fluctuations for U.S. sector index ETFs. By leveraging information-theoretic measures like mutual information and transfer entropy, we constructed threshold networks highlighting nonlinear dependencies between log returns and trading volume rate changes. We derived centrality measures and nod...

Parametric optimization solves a family of optimization problems as a function of parameters. It is a critical component in situations where optimal decision making is repeatedly performed for updated parameter values, but computation becomes challenging when complex problems need to be solved in real-time. Therefore, in this study, we present theo...

Machine learning has been widely used in the asset management industry to improve operations and make data-driven decisions. This article provides an overview of machine learning for asset management by presenting various machine learning models in the context of their applications, including general classification and regression, time-series forec...

The rise of FinTech has transformed financial services onto online platforms, yet stock investment recommender systems have received limited attention compared to other industries. Personalized stock recommendations can significantly impact customer engagement and satisfaction within the industry. However, traditional investment recommendations foc...

A neural networks-based stagewise decomposition algorithm called Deep Value Function Networks (DVFN) is proposed for large-scale multi-stage stochastic programming (MSP) problems. Traditional approaches such as nested Benders decomposition and its stochastic variant, stochas-tic dual dynamic programming (SDDP) approximates value functions as piecew...

This study used information theory and network theory to predict the fluctuations of currency values of the machine learning model. For experiments, we calculate the causal relationships between currencies using loarithmic return (log-return) and entropic value-at-risk (EVaR) values of gold price per troy ounce in 48 currencies over 25 years. To qu...

The collaborative filtering (CF) recommendation algorithm predicts the purchases of specific users based on their characteristics and purchase history. This study empirically analyzes the possibility of applying CF to the insurance industry using real customer data from South Korea. Using three different CF models, we examined the relevance of appl...

The mean-variance (MV) framework has been a fundamental tenet of investment management, yet it has been criticized for being too sensitive to parameter estimation errors. Hence, it is important to understand how the errors in parameters affect the MV framework. Although a number of researchers have studied how errors in parameters affect MV optimal...

A stagewise decomposition algorithm called value function gradient learning (VFGL) is proposed for large-scale multistage stochastic convex programs. VFGL finds the parameter values that best fit the gradient of the value function within a given parametric family. Widely-used decomposition algorithms for multistage stochastic programming, such as s...

In this study, we observed the changes in dietary patterns among Korean adults in the previous decade. We evaluated dietary intake using 24-h recall data from the fourth (2007–2009) and seventh (2016–2018) Korea National Health and Nutrition Examination Survey. Machine learning-based methodologies were used to extract these dietary patterns. Partic...

While portfolio optimization is generally based on the return and risk of a portfolio, goal-based investing primarily focuses on achieving financial goals of individuals, which has become a popular approach in personalized financial planning. While many long-term financial planning models use scenarios for representing uncertainty in future market...

We present a model based on trend filtering and regularization for performing factor analysis. Furthermore, the trend filtering method proposed in our model allows incorporating views represented as scenarios. Therefore, factor models can be optimized to explain not only trends of a given data series but also trends reflecting outlooks. As an appli...

Politically-themed stocks mainly refer to stocks that benefit from the policies of politicians. This study gave the empirical analysis of the politically-themed stocks in the Republic of Korea and constructed politically-themed stock networks based on the Republic of Korea’s politically-themed stocks, derived mainly from politicians. To select poli...

The mean-variance model is widely acknowledged as the foundation of portfolio allocation because it provides a framework for analyzing the tradeoff between risk and return for gaining diversification benefits. The shortcomings of the model may be just as widely recognized, but it is often the starting point for making asset allocation decisions. In...

The Korean National Pension Service (NPS) is a partially funded and defined-benefit system. Although the accumulated Fund of the NPS has been increased gradually, this large fund is concerned about depletion in the near future due to the unprecedented aging population and the low fertility rate. In this study, we have developed an asset-liability m...

While many individuals make investments to gain financial stability, most individual investors hold under-diversified portfolios that consist of only a few financial assets. Lack of diversification is alarming especially for average individuals because it may result in massive drawdowns in their portfolio returns. In this study, we analyze if it is...

Short-swing profit rule mandates insiders to disgorge short-term profits, thus preventing them from making short-swing trades. It can be imposed on investors who otherwise do not qualify as insiders, if they exercise certain shareholder engagements. This study provides a closed-form solution for the expected value of opportunity cost at the portfol...

We extend Merton’s framework by adopting stochastic volatility to propose an early warning indicator for banks’ credit risk. Bayesian inference is employed to estimate the parameters of Heston model. We provide empirical evidence and demonstrate the comparative strength of our risk measure over others.

In this paper, we propose a goal-based investment model that is suitable for personalized wealth management. The model only requires a few intuitive inputs such as size of wealth, investment amount, and consumption goals from individual investors. In particular, a priority level can be assigned to each consumption goal and the model provides a holi...

In investment management, especially for automated investment services, it is critical for portfolios to have a manageable number of assets and robust performance. First, portfolios should not contain too many assets in order to reduce the management fees, transaction costs, and taxes. Second, portfolios should be robust as investment environments...

Objective We analysed optimal nutrient levels using linear programming (LP) to reveal nutritional shortcomings of Korean dine-out meals and to stress the importance of fruits and dairy products for maintaining a healthy diet. Design LP models that minimize deviation from recommended nutrient values were formulated to analyse deficiency or excess o...

Smart beta, which accounts for rule-based factor-tilting strategies that fall between active and passive investment, has emerged as an alternative to active investment after its major decline since the global financial crisis. In spite of the smart beta's remarkable commercial prosperity, many experts in both industry and academia share some concer...

Robust optimization has become a widely implemented approach in investment management for incorporating uncertainty into financial models. The first applications were to asset allocation and equity portfolio construction. Significant advancements in robust portfolio optimization took place since it gained popularity almost two decades ago for impro...

The earliest documented analytical approach to portfolio selection is Markowitz’s mean–variance analysis, which attempts to find the portfolio with optimal performance by considering the tradeoff between return and risk. The performance of mean–variance analysis has been the subject of many studies and compared to other portfolio construction appro...

We propose an implementable portfolio performance evaluation procedure that compares a portfolio with respect to the portfolios constructed by an infinite number of Malkiel’s blindfolded monkeys, or equivalently the whole enumeration of all possible portfolios. We argue that this approach exhibits two main advantages. First, it does not require any...

Traditionally, the private wealth management industry has been dedicated to high-net-worth individuals due to expensive service costs. However, robo-advisors are making sudden rises during the on-going FinTech revolution. Robo-advisors aim to provide personalized wealth management services for everyone by using automated investment management algor...

Current market conditions pose new challenges for institutional investors. Traditional asset and liability models are struggling to meet investors’ needs due to poor performance of equity and bond markets. The move of portfolio allocation to alternative assets is evident. As a result, illiquidity issues and rebalancing difficulty arise. We propose...

This paper aims to provide a mathematical justification for asset allocation. Asset allocation is widely employed in practice, yet the question of its efficiency remains open. Asset allocation allows portfolio managers to concentrate on a relatively small number of assets, while security selection offers a bigger opportunity set for those who are c...

Mean-variance analysis is considered the foundation of portfolio selection. Among various attempts to address the limitations of the original model as formulated by Markowitz more than 60 years ago, one simple solution has been to impose constraints on weights in order to reduce efficient portfolios with extreme weights that may be caused by estima...

In this study, we propose a uniformly distributed random portfolio as an alternative benchmark for portfolio performance evaluation. The uniformly distributed random portfolio is analogous to an enumeration of all feasible portfolios without any prior on the market. Therefore, the relative ranking of a portfolio can be evaluated without peer group...

Mean-variance analysis is powerful for figuring out the optimal allocation of investments. The framework is straightforward, as it uses mean, variance, and covariance of asset returns for finding the trade-off between return and risk. Before discussing ways to improve the framework, this chapter focuses on the limitations of the use of variance for...

This chapter begins by describing the concept of robustness and then introduces the most widely recognized robust approaches for portfolio construction: Robust statistics, shrinkage estimation, Monte Carlo simulation (portfolio resampling), constraining portfolio weights, Bayesian approach (Black-Litterman model) and stochastic programming. The equ...

Identifying the factors in the equity market that drive returns and understanding the sensitivity of a portfolio to factor movements are extremely important for portfolio construction and risk management. If robust optimization has a significant impact on a portfolio's factor exposure, one cannot simply update the original portfolio to its robust c...

This chapter explains the construction of mean-variance and robust portfolios using optimization tools. The objectives in the chapter are to: solve the classical mean-variance problem using MATLAB, formulate robust counterparts as min-max problems with box and ellipsoidal uncertainty sets for the expected return of stocks, and explain how to constr...

This chapter discusses the three optimization tools that can help solve robust portfolio optimization: YALMIP, Robust Optimization Made Easy (ROME) and Advanced Integrated Multidimensional Modeling Software (AIMMS). YALMIP provides functions built precisely for robust optimization, and it also performs reformulation of an uncertain problem. ROME pr...

This chapter briefly reviews portfolio theory as formulated by Harry Markowitz in 1952. Portfolio theory explains how to construct portfolios based on the correlation of the mean, variance, and covariance of asset returns. Mean-variance analysis not only provides a framework for selecting portfolios, it also explains how portfolio risk is reduced b...

This chapter introduces robust optimization, and explains how uncertainty-averse decisions are a result of thinking about worst-case scenarios. It provides a brief introduction to convex optimization, and summarizes the different types of optimization problems. Based on the objective and constraint functions, optimization problems can be classified...

This chapter describes analyses on robust equity portfolios at the stock level. Observing how robust portfolio optimization allocates weights to individual stocks provide further understanding of robust portfolios in equities. Stock-level analyses focus on the following: composition based on investment styles, composition based on momentum, composi...

This chapter shows how to evaluate the performance of portfolios and also investigate how robust portfolios perform compared to other conventional approaches. The chapter introduces common measures of portfolio return and risk, demonstrates the implementation of performance measures using MATLAB and presents historical performance of various portfo...

Despite its shortcomings, the Markowitz model remains the norm for asset allocation and portfolio construction. A major issue involves sensitivity of the model’s solution to its input parameters. The prevailing approach employed by practitioners to overcome this problem is to use worst-case optimization. Generally, these methods have been adopted w...

The development of robo-advising allows investors to receive advice on investments that were considered too small by traditional human advisers. In this paper, the viability of robo-advising for individual investors is discussed by investigating one of the basic strategies of portfolio management, diversification. We formulate a mean-variance portf...

For portfolio management in the real-world, it is required that a portfolio has a manageable number of assets and stable performance. However, much research has pointed out that the Markowitz model, which is a classical model in portfolio theory, forms a portfolio with many different assets that may have unstable performance. Therefore, in this pap...

It is widely believed that the 1/n portfolio provides a good ex-post performance. Several studies have compared the 1/n portfolio with respect to a set of optimal mean-variance policies to prove or disprove the superiority of the 1/n portfolio. However, this approach is not likely to yield a definitive answer, since it provides only relative inform...

Many investors employ asset allocation, even though most are not really concerned about how their asset classification schemes affect investment performance. This article extensively examines the two most widely employed within-stock classifications: styles and industry classification. In order to explicitly measure current classifications’ perform...

Robust portfolio optimization has been developed to resolve the high sensitivity to inputs of the Markowitz mean-variance model. Although much effort has been put into forming robust portfolios, there have not been many attempts to analyze the characteristics of portfolios formed from robust optimization. We investigate the behavior of robust portf...

Robust portfolios reduce the uncertainty in portfolio performance. In particular, the worst-case optimization approach is based on the Markowitz model and form portfolios that are more robust compared to mean–variance portfolios. However, since the robust formulation finds a different portfolio from the optimal mean–variance portfolio, the two port...

Asset allocation among diverse financial markets is essential for investors especially under situations such as the financial crisis of 2008. Portfolio optimization is the most developed method to examine the optimal decision for asset allocation. We employ the hidden Markov model to identify regimes in varied financial markets; a regime switching...

In recent years, there has been an increasing interest in hyperspherical caps from machine learning domain. Concise formulas for the area of a hyperspherical cap are now available and being beneficial for researchers, but there is not one for the intersection of two hyperspherical caps in spite of its large potential in application. This paper prov...

In spite of their importance, third or higher moments of portfolio returns are often neglected in portfolio construction problems due to the computational difficulties associated with them. In this paper, we propose a new robust mean–variance approach that can control portfolio skewness and kurtosis without imposing higher moment terms. The key ide...

In this paper, we find that the robust portfolios constructed from worst-case approaches systematically bet more on the factors in an asset universe composed of equities, fixed incomes, and commodities. This generalizes the findings that the robust equity portfolios are more tilted towards Fama–French factors than mean–variance portfolios. In addit...

Robust portfolios resolve the sensitivity issue identified as a concern in implementing mean–variance analysis. Because robust approaches are not widely used in practice due to a limited understanding regarding the portfolios constructed from these methods, we present an analysis of the composition of robust equity portfolios. We find that compared...

This study presents an application of stochastic model for limit order book (LOB) dynamics to Korean Stock Index Futures (KOSPI 200 Futures). Since KOSPI 200 futures market is widely known as one of the most liquid markets in the world, direct application of an existing model is hardly possible. Therefore, we modified an existing model to successfu...

Most of previous work on robust equity portfolio optimization has focused on its formulation and performance. In contrast, in this paper we analyze the behavior of robust equity portfolios to determine whether reducing the sensitivity to input estimation errors is all robust models do and investigate any side-effects of robust formulations. Therefo...

Robust models have a major role in portfolio optimization for resolving the sensitivity issue of the classical mean–variance model. In this paper, we survey developments of worst-case optimization while focusing on approaches for constructing robust portfolios. In addition to the robust formulations for the Markowitz model, we review work on derivi...