• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2003년도 춘계학술대회
• /
• pp.329-337
• /
• 2003

Stock market prediction is regarded as a challenging task of financial time-series prediction. There have been many studies using artificial neural networks in this area. Recently, support vector machines (SVMs) are regarded as promising methods for the prediction of financial time-series because they me a risk function consisting the empirical ewer and a regularized term which is derived from the structural risk minimization principle. In this study, I apply SVM to predicting the Korea Composite Stock Price Index (KOSPI). In addition, this study examines the feasibility of applying SVM in financial forecasting by comparing it with back-propagation neural networks and case-based reasoning. The experimental results show that SVM provides a promising alternative to stock market prediction.

• 한국인공지능학회지
• /
• 제7권1호
• /
• pp.23-59
• /
• 2019

The purpose of this study is to develop a measure of service quality in the traditional market by examining previous research on the service quality of the traditional market studied so far. After defining basic concepts through definition of traditional market and existing studies, 5 categories of configuration items for SERVQUAL measurement in traditional market were made up based on existing researches related to definition of service quality and service quality of traditional market. A survey was conducted on the items that fit the intention of this study and various statistical analyzes were conducted. Statistical analysis was performed using SPSS 22.0 and AMOS 22.0. The reliability of the items was measured by the reliability test, and the predictability and accuracy of the items were examined. The validity of the measured variables was verified through confirmatory factor analysis. Reliability, empathy, responsiveness, certainty, and tangibility were the most important factors in this study. Responsiveness factors include communication, time reduction, real time, promptness. Assurance factors include the assurance of delivery, prompt answers, product knowledge items. Tangibility factors include, convenient device systems, location information, presence as a fact, and as a result, the latest modern items are adopted. The quality of service in the traditional market developed in this study was found to be good in reliability and validity test. Confirmatory factor analysis result using structural equation model also met the conformity index standard. If service satisfaction is measured based on this research, basic data can be presented to policy makers who implement policies on traditional markets to make the right decisions. In addition, it will be able to provide traditional market operators with operational strategy and marketing data. In the future, based on the traditional market service quality scale developed in this study, it is necessary to grasp the factors to be continuously managed to improve the service quality of the traditional market, user satisfaction, and intention to use.

• Asia pacific journal of information systems
• /
• 제15권4호
• /
• pp.165-192
• /
• 2005

This study introduces how to estimate the monetary value of intellectual capital of a public research institute by incorporating a non-market valuation technique, the choice experiments(CE). CE is a survey-based environmental valuation technique that has increasingly been popular over the last decade. The members of institute E, a typical type of public research institutes in Korea, were surveyed, before the data were fit to the conditional logit and mixed logit models. The total value of the institute's intellectual capital was estimated at approximately W3,377 billion for the year 2003. The institute's human, structural and relational capitals that comprise the intellectual capital were estimated at W18.7 billion, W10.7 billion and W4.4 billion respectively, for each of the components' index values improving by 1%. The human capital was placed a higher value than the other two. The study also shows that CE is a flexible technique that enables the researcher to estimate the monetary value of the intellectual capital whatever the index values of the component capitals and to interpret model estimation results more in depth by incorporating the mixed logit, a state-of-the-art discrete choice model, than the conventional conditional logic.

• 유통과학연구
• /
• 제14권12호
• /
• pp.79-87
• /
• 2016

Purpose - Since 1990, the sudden stop in capital flows has caused the economic crisis. The purpose of this research is to suggest the policy measures to mitigate the risk of the sudden stop in capital flows. To this end, we examine the theoretical framework and analyze the case study for countries which are faced with the sudden stop. Also we examine the structural problems of the foreign exchange market in Korea and derive the policy implications to prevent the sudden stop. Research design, data, and methodology - The criteria of whether the sudden stop in capital flows occurs are based upon Calvo et al. (2008). In case the proxy variable for the balance of capital account decreases from the average by over twice standard deviation, we determine that the sudden stop occurs for that country. The sample period is from January 1990 to December 2008, as in Calvo (2014). The sample countries are 17 developed countries and 19 emerging market countries, which are different from those of the previous papers as Agosin and Huaita (2012), and Calvo (2014). When the exchange market pressure index(EMPI) is deviated from the average by over three times standard deviation, we determine that the foreign exchange market is unstable for that country. Results - We find that the characteristics of the sudden stop in capital flows are the bunching or contagion among countries, the rapid drop in real effective exchange rate, and the huge decrease in foreign exchange reserves. Many countries tried to increase foreign exchange reserves and regulate capital flows. Also the foreign exchange market in Korea are found to be the volatile exchange rate, the vulnerable external debt and careless management of the foreign exchange derivatives transaction risk. Conclusions - To lessen the risk in the sudden stop of capital flows, this research suggests the some useful policy measures. To enhance the foreign exchange and distribution market stability, we should improve the price mechanism of exchange rate, hold the appropriate level of foreign exchange reserves, prevent excessive inflows of foreign exchange and promote sound transactions of foreign exchange derivatives.

• Asian Journal of Innovation and Policy
• /
• 제12권3호
• /
• pp.323-344
• /
• 2023

This study investigates the resilient structural characteristics of a region by assessing the impact of the financial crisis. Utilizing panel data at the prefecture level for metropolitan cities across pre-shock (2006-2008), shock (2009), and post-shock (2010-2019) periods, we calculated an employment resilience index by combining the resistance and recovery indices. The panel logit regression measures the influences of the region's industrial structure and external economic factors in response to the global financial crisis. The results revealed that the diversity index of industries contributed to the post-shock recovery bounce-back. Additionally, the presence of large firms and industrial clusters within the region positively contributed to economic resilience. The specialization and the proportion of manufacturing industries showed negative effects, suggesting that regions overly reliant on manufacturing-centered specialization might be vulnerable to external shocks. Furthermore, excessive capital outflows for market expansion were found to have a detrimental impact on regional economic recovery.

• 유통과학연구
• /
• 제11권10호
• /
• pp.73-79
• /
• 2013

Purpose - This study examines the impact of oil price volatility on economic activities in Korea. The new millennium has seen a deregulation in the crude oil market, which invited immense capital inflow into Korea. It has also raised oil price levels and volatility. Drawing on the recent theoretical literature that emphasizes the role of volatility, this paper attends to the asymmetric changes in economic growth in response to the oil price movement. This study further examines several key macroeconomic variables, such as interest rate, production, and inflation. We come to the conclusion that oil price volatility can, in some part, explain the structural changes. Research design, data, and methodology - We use two methodological frameworks in this study. First, in regards to the oil price uncertainty, we use an Exponential-GARCH (Exponential Generalized Autoregressive Conditional Heteroskedasticity: EGARCH) model estimate to elucidate the asymmetric effect of oil price shock on the conditional oil price volatility. Second, along with the estimation of the conditional volatility by the EGARCH model, we use the estimates in a VECM (Vector Error Correction Model). The study thus examines the dynamic impacts of oil price volatility on industrial production, price levels, and monetary policy responses. We also approximate the monetary policy function by the yield of monetary stabilization bond. The data collected for the study ranges from 1990: M1 to 2013: M7. In the VECM analysis section, the time span is split into two sub-periods; one from 1990 to 1999, and another from 2000 to 2013, due to the U.S. CFTC (Commodity Futures Trading Commission) deregulation on the crude oil futures that became effective in 2000. This paper intends to probe the relationship between oil price uncertainty and macroeconomic variables since the structural change in the oil market became effective. Results and Conclusions - The dynamic impulse response functions obtained from the VECM show a prolonged dampening effect of oil price volatility shock on the industrial production across all sub-periods. We also find that inflation measured by CPI rises by one standard deviation shock in response to oil price uncertainty, and lasts for the ensuing period. In addition, the impulse response functions allude that South Korea practices an expansionary monetary policy in response to oil price shocks, which stems from oil price uncertainty. Moreover, a comparison of the results of the dynamic impulse response functions from the two sub-periods suggests that the dynamic relationships have strengthened since 2000. Specifically, the results are most drastic in terms of industrial production; the impact of oil price volatility shocks has more than doubled from the year 2000 onwards. These results again indicate that the relationships between crude oil price uncertainty and Korean macroeconomic activities have been strengthened since the year2000, which resulted in a structural change in the crude oil market due to the deregulation of the crude oil futures.

• 응용통계연구
• /
• 제24권2호
• /
• pp.239-251
• /
• 2011

국가 간 경기 동조화 현장은 한 나라의 경기변동이 긴 시차를 두지 않고 관련국의 경기와 금융시장의 변동으로 직결되면서 동반 이동이 어떤 형태를 보이는지에 대한 통계적 분석의 필요성이 중대하고 있다. 본 연구에서는 동조화 현상과 파급효과에 대한 견인차 가설에 입각해 규모가 큰 경제가 작은 규모 국가의 변동을 유발하는지의 여부와 성향을 파악하고자 한 미 양국의 현재와 미래의 경기동향 및 주식시장 간 장기적 역학관계와 구조적 성향에 대해 통계적으로 검정해 보았다. 국내 주식시장에는 국내 경기동향이나 미국의 현 경기여건보다는 미국의 미래 경기전망이 유의한 영향을 미치는 것으로 나타났다.

• The Journal of Asian Finance, Economics and Business
• /
• 제8권4호
• /
• pp.113-124
• /
• 2021

The study investigates how financial markets respond to a shock to tone and semantic similarity of the Bank of Thailand press releases. The techniques in natural language processing are employed to quantify the tone and the semantic similarity of 69 press releases from 2010 to 2018. The corpus of the press releases is accessible to the general public. Stock market returns and bond yields are measured by logged return on SET50 and short-term and long-term government bonds, respectively. Data are daily from January 4, 2010, to August 8, 2019. The study uses the Structural Vector Auto Regressive model (SVAR) to analyze the effects of unanticipated and temporary shocks to the tone and the semantic similarity on bond yields and stock market returns. Impulse response functions are also constructed for the analysis. The results show that 1-month, 3-month, 6-month and 1-year bond yields significantly increase in response to a positive shock to the tone of press releases and 1-month, 3-month, 6-month, 1-year and 25-year bond yields significantly increase in response to a positive shock to the semantic similarity. Interestingly, stock market returns obtained from the SET50 index insignificantly respond to the shocks from the tone and the semantic similarity of the press releases.

• 지능정보연구
• /
• 제22권4호
• /
• pp.19-43
• /
• 2016

이 연구에서 Asia 금융 위기의 원인을 고찰하여 보고, European Monetary Systems의 금융 위기와 비교하여 본다. Asian 신흥 국가들은 1997년도에 금융 위기를 경험하였고, European Monetary Systems의 국가들도 1992년도에 동일한 경험을 하였다. 또한, 중남미의 신흥 경제국가인 Mexico 역시 1994년에 금융위기를 겪었다. 이 연구의 목적은 이들 금융위기의 내면을 고찰하고 그 결과로부터 일반화된 법칙을 추출하는 것이다. 이 연구에서는 금융위기를 경험한 한국과 영국과 멕시코를 각각 세가지 다른 모형으로 연구하고 비교하였다. 이 접근 방법은 체계적인 조사를 통하여 세 국가의 차이점을 보여주고 또한 공통적인 내재 요인을 관찰한다. 이전의 많은 연구 방법들은 대부분 선형 회귀식을 통한 causal model에 초점을 맞추고 있지만, 이러한 선형 회귀 모형의 약점을 보완하여서 현실에 산재하며 존재하는 비 선형의 문제를 해결하기 위하여 또 다른 방법을 제안하여 본다. 이 연구에서 사용한 구조 방정식(Structural Equation Model) 모형은 현실로부터 원인을 추출하고 분석하는 연구에 적합하며, 신경망(Artificial Neural Network) 모형은 선형모형의 단점을 보완하여서 비 선형 요인을 설명해 준다. 구조방정식 모형에 적용하기 위하여서 LISREL(LInear Structural RELationship)을 사용하였다. LISREL은 확인적 요인분석과 계량경제학에서 개발된 연립방정식모델에 토대를 둔 다중회귀분석 및 경로분석 등이 결합된 성격을 갖는 방법론으로 다양한 연구에 적용된다. 또한 인공지능(Artificial Intelligence) 기법 중의 하나인 신경망 모형은 선형회귀 분석과 다른 형태의 결과를 도출한다. 세가지 방법론의 우수성을 비교하기 위하여 Hit ratio를 각 국가/ 각 방법론 별로 구분하여서 비교한 결과 다른 방법론 보다 신경망이 더 좋은 성과를 나타내고 있는 것을 확인할 수 있었다. 세가지 방법론에 각각 일반적인 환율 예측에 사용되는 변수를 사용하였다. 소비자 물가지수(Consumer Price Index), 국내총생산(Gross Domestic Product), 이자율(Interest rate), 주가지수(Stock Index), 경상수지(Current Account), 외환보유고(Foreign Reserves)의 6가지 변수를 이용하여서 환율을 예측하여서 급격한 환율 변화로 초래되는 경제위기를 예측하려고 하였다. 각각의 국가의 데이터는 대한민국은 1991년부터 1999년까지, 영국은 1986년부터 1995년까지, 멕시코는 1988년부터 1998년까지의 기간을 정하여서 시계열자료를 분기별로 사용하였다. 각각의 데이터는 Data Stream과 한국은행(Bank of Korea)의 데이터를 이용하여서 분석하였다. 선형회귀방정식을 이용한 분석과 구조방정식인 LISREL을 이용한 분석은 각각 Hit ratio가 국가별로 순위가 변동되기도 하였으나, 인공지능 방법론인 인공신경망의 경우는 모든 국가에서 가장 좋은 예측 결과를 나타내고 있었다. 이 논문은 환율의 변동에 대한 다양한 예측 모형을 비교하고 평가하여서 연구에서 제시하는 개념을 검토하였다는 점에서 의의를 갖는다.

• 국제지역연구
• /
• 제14권3호
• /
• pp.62-84
• /
• 2010

본 논문은 중국, 홍콩, 대만 주식시장들 사이의 동태적 상호의존성을 연구한다. 이를 위하여 아시아 금융위기가 그러한 상호의존성의 구조전환점인지를 검토하고, 이를 아시아 금융위기를 기준으로 세 가지 분석기간을 설정하여 수익률과 변동성의 정보전이효과를 분석한다. 전체기간을 대상으로 한 실증분석 결과 세 시장 수익률 평균과 비대칭 변동성 사이에 정보전이효과가 유의하게 존재한다는 증거가 발견되었다. 이는 세 시장 간에 정보전이와 비대칭적 변동성이 존재한다는 것을 암시한다. 또 수익률 평균과 비대칭 변동성 사이에 존재하는 정보전이효과의 크기가 금융위기 이후 증가한 것으로 나타났다. 이러한 사실은 아시아 금융위기 이후 중국, 홍콩, 대만 주식시장의 통합이 더 강화된 것을 의미한다. 특히 변동성 정보전이효과의 비대칭성이 금융위기 이후 더 심화된 것으로 나타났다. 이러한 사실은 긍정적 충격보다 부정적 충격이 대중국 주식시장 변동성에 미치는 영향이 금융위기 이후 더 심화된 것을 의미한다. 결론적으로 아시아 금융위기가 중국, 홍콩, 대만 주식시장의 정보전이와 비대칭성을 심화시킨 것으로 판단된다.