This page has only limited features, please log in for full access.
Portfolio management is a series of processes that maximize returns and minimize risk by allocating assets efficiently. Along with the developments in machine learning technology, it has been studied to apply machine learning methods to prediction-based portfolio management. However, such methods have a few limitations. First, they do not consider the relations between assets for the prediction. In addition, the studies commonly focus on the prediction accuracy, neglecting the construction of portfolios. Furthermore, the methods have usually been evaluated with index data, which hardly represent actual prices to buy or sell an asset. To overcome these problems, Exchange Traded Funds (ETFs) are employed for base assets for the evaluation, and we propose a two-stage deep learning framework, called Grouped-ETFs Model (GEM), with a joint cost function. The GEM is designed to learn the features of inter-asset and groups in each stage. Also, the proposed joint cost can consider relative returns for the training while the relative returns are a crucial factor to construct a portfolio. The results of a rigorous evaluation with global ETF data indicate that the proposed GEM with the joint cost outperforms the equally weighted portfolio and the ordinary deep learning model by 33.7% and 30.1%, respectively. An additional experiment using sector ETFs verifies the generality of the proposed model where the results accord with those of the previous experiment.
Hyungbin Yun; Minhyeok Lee; Yeong Seon Kang; Junhee Seok. Portfolio management via two-stage deep learning with a joint cost. Expert Systems with Applications 2019, 143, 113041 .
AMA StyleHyungbin Yun, Minhyeok Lee, Yeong Seon Kang, Junhee Seok. Portfolio management via two-stage deep learning with a joint cost. Expert Systems with Applications. 2019; 143 ():113041.
Chicago/Turabian StyleHyungbin Yun; Minhyeok Lee; Yeong Seon Kang; Junhee Seok. 2019. "Portfolio management via two-stage deep learning with a joint cost." Expert Systems with Applications 143, no. : 113041.
Addressing the fact that there are few studies exploring the relationship between board characteristics and corporate social responsibility (CSR) in non-Western contexts, this study examines the relationship in South Korean corporate contexts. We concentrate on foreign directors as a board attribute, which is reported as a remarkable change in Korean corporate boards, and propose that foreign directors have different impacts on CSR investment depending on their nationality (Anglo-Americans vs. non-Anglo-Americans) and director types (insiders vs. outsiders). In detail, the presence of directors from Anglo-American countries (e.g., the United States, the United Kingdom) decreases firms’ CSR involvement, whereas the presence of directors from non-Anglo-American countries (e.g., France, Germany) increases firms’ CSR involvement. Moreover, the effects of Anglo-Americans on CSR are strengthened when they are inside (rather than outside) directors. Empirical analyses using a sample of 1828 Korean firms from 2002 to 2015 provide evidence to support the predictions. This study theoretically contributes to CSR and corporate governance literature in that it sheds light on the CSR in non-Western companies and reveals varied effects of foreign directors contingent upon their individual attributes. It also has practical implications for policymakers and corporate managers by providing insights of the changes generated by foreign members in a boardroom.
Yeong Seon Kang; Eunji Huh; Mi-Hee Lim. Effects of Foreign Directors’ Nationalities and Director Types on Corporate Philanthropic Behavior: Evidence from Korean Firms. Sustainability 2019, 11, 3132 .
AMA StyleYeong Seon Kang, Eunji Huh, Mi-Hee Lim. Effects of Foreign Directors’ Nationalities and Director Types on Corporate Philanthropic Behavior: Evidence from Korean Firms. Sustainability. 2019; 11 (11):3132.
Chicago/Turabian StyleYeong Seon Kang; Eunji Huh; Mi-Hee Lim. 2019. "Effects of Foreign Directors’ Nationalities and Director Types on Corporate Philanthropic Behavior: Evidence from Korean Firms." Sustainability 11, no. 11: 3132.
Our objective in this study is to understand how adolescents respond to the food industry’s corporate social responsibility (CSR) activities, especially the effects of such activities on consumers’ emotional responses, perceived authenticity, and attitudes toward the company. Understanding which types of CSR actions most influence adolescents is important for managers. This study examines adolescents’ responses to three types of CSR actions (career-related, environment-related, and wellbeing-related) across two types of products (unhealthy and healthy foods). We find that CSR actions related to career issues have the greatest effects on adolescents’ emotional responses, perceived authenticity,and attitudes toward a company under the condition of healthy food products. In other words, when a healthy food company offers a career-related CSR program, adolescents have better responses than when an unhealthy food company offers the same CSR program.
Mi-Hee Lim; Yeong Seon Kang; Yura Kim. Effects of Corporate Social Responsibility Actions on South Korean Adolescents’ Perceptions in the Food Industry. Sustainability 2017, 9, 176 .
AMA StyleMi-Hee Lim, Yeong Seon Kang, Yura Kim. Effects of Corporate Social Responsibility Actions on South Korean Adolescents’ Perceptions in the Food Industry. Sustainability. 2017; 9 (2):176.
Chicago/Turabian StyleMi-Hee Lim; Yeong Seon Kang; Yura Kim. 2017. "Effects of Corporate Social Responsibility Actions on South Korean Adolescents’ Perceptions in the Food Industry." Sustainability 9, no. 2: 176.
Mutual information, a general measure of the relatedness between two random variables, has been actively used in the analysis of biomedical data. The mutual information between two discrete variables is conventionally calculated by their joint probabilities estimated from the frequency of observed samples in each combination of variable categories. However, this conventional approach is no longer efficient for discrete variables with many categories, which can be easily found in large-scale biomedical data such as diagnosis codes, drug compounds, and genotypes. Here, we propose a method to provide stable estimations for the mutual information between discrete variables with many categories. Simulation studies showed that the proposed method reduced the estimation errors by 45 folds and improved the correlation coefficients with true values by 99 folds, compared with the conventional calculation of mutual information. The proposed method was also demonstrated through a case study for diagnostic data in electronic health records. This method is expected to be useful in the analysis of various biomedical data with discrete variables.
Junhee Seok; Yeong Seon Kang. Mutual Information between Discrete Variables with Many Categories using Recursive Adaptive Partitioning. Scientific Reports 2015, 5, 10981 .
AMA StyleJunhee Seok, Yeong Seon Kang. Mutual Information between Discrete Variables with Many Categories using Recursive Adaptive Partitioning. Scientific Reports. 2015; 5 (1):10981.
Chicago/Turabian StyleJunhee Seok; Yeong Seon Kang. 2015. "Mutual Information between Discrete Variables with Many Categories using Recursive Adaptive Partitioning." Scientific Reports 5, no. 1: 10981.