개인투자자의 주식투자 성과 분석

변영훈 한국재무관리학회 2005 財務管理硏究 Vol.22 No.2

개인투자자 10,000명의 1998년부터 2003년까지 6년간의 거래자료와 잔고자료를 분석한 결과, 개인투자자들은 총수익률(gross return) 기준으로 연간 12.3%의 수익률을 실현한 것으로 나타났다. 동기간에 거래소 시장의 가치가중평균수익률은 13.6%였으며 코스닥 시장을 포함하는 종합시장수익률은 9.7%를 기록하였다. 그러나 거래비용을 고려한 순수익률(net return)은 연간 8.3%로 하락하여 시장수익률보다 크게 낮은 것으로 나타났는데, 연간 270%가 넘는 거래 회전율이 투자성과에 부정적인 영향을 미친 탓이라 할 수 있다. 특히 잔고규모별 초과수익률 분석에서는 투자금액이 상위 20%에 속하는 투자자들은 시장수익률과 비슷한 수준의 수익률을 얻은 반면, 나머지 80%의 투자자들은 시장수익률과 커다란 차이를 보였는데, 왜 많은 개인 투자자들이 시장을 떠나는지를 확인시켜 주는 결과이다. 특기할 사항은 과잉확신으로 인해 거래량이 증가하고 기대효용은 낮아진다는 과잉확신 모형을 지지한다는 점이다. 분석대상인 개인투자자들은 고 베타 주식과 소형주와 가치주를 선호하는 것으로 나타났다. We analyze trade and balance records of 10,000 stock investment accounts of individual investors for the period of 1998 to 2003. Individual investors earn an annual gross return of 12.3% while the KOSPI and the value weighted composite including KOSDAQ stocks yield 13.6% and 9.7% respectively during the same period. Net return performance is 8.3%, a drop of 5.3% mainly due to heavy trading. Individual investors’ annual turnover amounts to over 270 percent. In an analysis of groups formed on the month’s end position value, the performance of the top quintile is found comparable to the market while the rest yield significantly lower risk-adjusted returns than the market. We also find evidence rejecting the rational expectation model while supporting the overconfidence hypothesis which states overconfidence leads to a higher level of trading, resulting in poor performance. Individuals tilt their stock investment toward high-beta, small, and value stocks.

쿼드로터 UAV의 PPO 기반 모델 참조 추종 제어

변영훈,김한솔 제어·로봇·시스템학회 2022 제어·로봇·시스템학회 논문지 Vol.28 No.11

This paper presents a proximal policy optimization (PPO)-based model reference tracking controller design for quadrotor unmanned aerial vehicles (UAVs). First, the quadrotor UAV is divided into the attitude and position systems, and each system is expressed as a nonlinear state-space equation. Thereafter, we design a linear reference model, which is used to derive the tracking error dynamics. The proposed neural network model implements a PPO-based reinforcement learning algorithm consisting of an actor approximating a policy and a critic corresponding to a state-value function. The actor receives the state variables of the tracking error dynamics as input and outputs the thrust values to be applied to each axis. Further, we decentralize the PPO-based controller into the attitude and position controllers, which are then trained separately. For learning purposes, we implement an environment code that expresses the tracking error dynamics of the quadrotor by extending the OpenAI Gym environment. Finally, a simulation example is provided to show the position tracking performances of up to 0.0166 m and 0.0254 m for the horizontal and vertical axes, respectively. .

Clinical features and risk factors for missed stroke team activation in cases of acute ischemic stroke in the emergency department

변영훈,홍성엽,우선희,김현정,정시경 대한응급의학회 2018 대한응급의학회지 Vol.29 No.5

Objective: Acute ischemic stroke (AIS) requires time-dependent reperfusion therapy, and early recognition of AIS is important to patient outcomes. This study was conducted to identify the clinical features and risk factors of AIS patients that are missed during the early stages of diagnosis. Methods: We retrospectively reviewed AIS patients admitted to a hospital through the emergency department. AIS patients were defined as ischemic stroke patients who visited the emergency department within 6 hours of symptom onset. Patients were classified into two groups: an activation group (A group), in which patients were identified as AIS and the stroke team was activated, and a non-activation group (NA group), for whom the stroke team was not activated. Results: The stroke team was activated for 213 of a total of 262 AIS patients (81.3%), while it was not activated for the remaining 49 (18.7%). The NA group was found to be younger, have lower initial National Institutes of Health Stroke Scale scores, lower incidence of previous hypertension, and a greater incidence of cerebellum and cardio-embolic infarcts than the A group. The chief complaints in the A group were traditional stroke symptoms, side weakness (61.0%), and speech disturbance (17.8%), whereas the NA group had non-traditional symptoms, dizziness (32.7%), and decreased levels of consciousness (22.4%). Independent factors associated with missed stroke team activation were nystagmus, nausea/vomiting, dizziness, gait disturbance, and general weakness. Conclusion: A high index of AIS suspicion is required to identify such patients with these findings. Education on focused neurological examinations and the development of clinical decision tools that could differentiate non-stroke and stroke are needed.

기업인수.합병공시 전후의 매수.매도가격차이 움직임에 대한 실증적 연구

변영훈,Byeon, Yeong-Hun 한국재무관리학회 1995 財務管理硏究 Vol.12 No.2

본 연구에서는 기업인수 합병공시 전후에 관찰되는 매수 매도가격차이중에서 정보비용부분의 움직임을 살펴봄으로써 정보비대칭하의 시장미시구조이론을 검증하였다. 공시일을 예측할 수 없는 기업인수 합병공시를 대상으로 함으로써 효율적시장가설의 검증을 병행하는데 본 연구의 의의가 있다. 검증의 결과는 시장미시구조이론과 효율적시장가설을 지지한다. 공시전 전체기간에 대한 분석에서는 스프레드의 증가가 없었으나 부분기간에 대한 분석에서 스페셜리스트가 스프레드를 증가시키는 것을 확인하였다. 스프레드의 증가는 공시 3일전과 4일전에 나타났으며 이는 정보거래자에 대한 손실을 피하기 위하여 스프레드를 증가시킨다는 이론의 예측과 일치하는 증거이다. 그러나 정보누출과 이의 감지에는 시간차이가 존재하였다. 우호적공개매수와 적대적공개매수의 비교분석에서도 기업인수의 실현여부와 관련된 정보비대칭 현상에 대해 이론의 예측과 일치하는 결과를 얻었다.

CLS 토큰으로 초기화된 양방향 LSTM을 이용한 DPT 기반 단안 깊이 추정 모델의 성능 향상

변영훈,이민우,김한솔,이정욱,김재훈 제어·로봇·시스템학회 2023 제어·로봇·시스템학회 논문지 Vol.29 No.12

Monocular depth estimation (MDE) is a computer vision task that estimates the depth value for each pixel in a single monocular RGB image. Given its inherently ill-posed nature, recent studies have focused on data-driven learning-based methods, particularly deep learning techniques based on convolutional neural networks. Building on the success of vision transformers (ViTs), ViT-based models have gained widespread attention for MDE applications. To enhance the performance of MDE, we introduce an intermediate bi-directional long-short term memory (BiLSTM) module, a common component in natural language processing, to learn correlations between each token in a sequence both in the forward and backward directions. Given that transformers naturally encode sequential data, leveraging BiLSTMs to learn the correlations between each token in both directions is a reasonable approach. In addition, we enhance the weight initialization method for BiLSTMs by employing the CLS tokens of transformers, which was originally developed for natural language processing, to make it applicable for processing two-dimensional image data. Subsequently, we developed an MDE model by inserting a BiLSTM module initialized using our method between the encoder and decoder of a DPT-based architecture. Our experiments reveal notable results, with our model achieving 0.913 in δ_1 and 0.097 in AbsRel for the NYU Depth V2 dataset. These results demonstrate a significant improvement over the previous models.

Proposal for Light Shelf System that Applies Biomimicry for Lighting Energy Conservation

변영훈,류리,김용성 사단법인 미래융합기술연구학회 2019 아시아태평양융합연구교류논문지 Vol.5 No.2

Various methods are being proposed to reduce the usage of lighting energy in buildings. However, those methods are not being applied practically due to issues such as external environmental factors. Therefore, this study analyzed the existing lighting energy conservation systems to propose a system that applies biomimicry as a method to supplement vulnerabilities. As a method of analysis, the concept and cases of light shelf systems. second, study on the concept of biomimicry and life forms evolved by adapting to the external environment. third a triple light shelf system with applied biomimicry is proposed and the architectural application method is analyzed and studied. As a result a triple light shelf system applying biomimicry is proposed. In conclustion the existing light shelf system cannot respond to the external environment. Meanwhile, the triple light shelf system with applied biomimicry which imitates life forms that adapt and evolve according to the external environment can respond immediately to the outside environment of the building, and it is expected to establish a pleasant environment for people indoors as well as to bring lighting energy reduction effects in the building

드론 제어를 위한 심층 강화 학습의 최신 연구 사례

변영훈,김한솔 제어로봇시스템학회 2022 제어로봇시스템학회지 Vol.28 No.4

팽대부 주위암 환자의 임상적 고찰

변영훈,홍성화,이상목 대한외과학회 1996 대한외과학회 학술대회 초록집 Vol.1996 No.5

효율적 메모리 관리를 통한 모바일 CNN 가속기의 최적화

변영훈,조준서,이영주 대한전자공학회 2018 전자공학회지 Vol.45 No.1

Organization

변영훈 ( Byun Young Hoon ) 한국기초조형학회 2018 한국기초조형학회 작품집 Vol.2018 No.2