• 아태비즈니스연구
• /
• 제13권3호
• /
• pp.183-208
• /
• 2022

Purpose - The main purpose of this study is to widely investigate the impact of recent pandemic crises on the synchronization of the world capital markets through 25 stock indices from major developed countries. Design/methodology/approach - This study collects 25 stock indices from major developed countries and the time period is between January 5, 2001 and February 24, 2022. The data sets used in the study include finance.yahoo.com and Investing.com.. The Granger causality analysis, unit-root test, VAR analysis, and forecasting error variance decomposition were hired in order to analyze the data. Findings - First, there are significant inter-relations among 25 countries around recent major pandemic crises(such as SARS, A(H1N1), MERS, and COVID19), which is consistent result with previous literature. Second, COVID19 shows much stronger impact on the world-wide synchronization than other pandemics. Third, the return volatility of each stock market varies, unit root tests show that daily stock index data are unstable while daily stock index returns are stable, and VAR(Vector Auto Regression) analyses presents significant inter-relations among 25 capital markets. Fourth, from the impulse response function analyses, we find that each market affects the other markets for short term periods, about 2~4 days, and no long term effect was not found. Fifth, Granger causality tests show one-side or two-sides synchronization between capital markets and we estimate, through forecasting error variance decomposition method, that the explanatory portions of each capital market on other markets vary from 10 to 80%. Research implications or Originality - The above results all together show that pandemic crises have strong effects on the synchronization of world capital markets and imply that these synchronizations should be carefully considered both in the investment decisions by individual investors and in the financial and economic policies by governments.

• 한국빅데이터학회지
• /
• 제3권1호
• /
• pp.13-20
• /
• 2018

주식 시장에 대한 예측은 오랜 기간 많은 이들의 꿈이었다. 하지만 수많은 노력에도 불구하고 주식 시장을 정확하게 예측하기란 쉬운 일이 아니었다. 본 연구는 주식 시장의 방향성에 주목하여 이 방향성을 예측할 수 있는 감성사전을 구축하는 새로운 방법을 제시한다. 이를 위해 2015년 1월 1일부터 2017년 12월 31일까지 3년간의 증시 뉴스 25,000여 건의 데이터를 수집하여, 문맥을 고려하기 위한 Word2Vec을 적용하였다. 이를 바탕으로 뉴스에 감성분석을 실시하여 KOSPI 종가 지수를 예측해 보았다.

• 대한임베디드공학회논문지
• /
• 제17권1호
• /
• pp.51-58
• /
• 2022

Various studies have been conducted from the past to the present because stock price forecasts provide stability in the national economy and huge profits to investors. Recently, there have been many studies that suggest stock price prediction models using various input data such as macroeconomic indicators and emotional analysis. However, since each study was conducted individually, it is difficult to objectively compare each method, and studies on their impact on stock price prediction are still insufficient. In this paper, the effect of input data currently mainly used on the stock price is evaluated through the predicted value of the deep learning model and the error rate of the actual stock price. In addition, unlike most papers in emotional analysis, emotional analysis using the news body was conducted, and a method of supplementing the results of each emotional analysis is proposed through three emotional analysis models. Through experiments predicting Microsoft's revised closing price, the results of emotional analysis were found to be the most important factor in stock price prediction. Especially, when all of input data is used, error rate of ensembled sentiment analysis model is reduced by 58% compared to the baseline.

• Communications for Statistical Applications and Methods
• /
• 제31권5호
• /
• pp.535-556
• /
• 2024

Measuring stock market volatility and its determinants is critical for stock market participants, as volatility spillover effects affect corporate performance. This study adopted a novel approach to analysing and implementing GARCH-MIDAS modelling methods. The classical GARCH as a benchmark and the univariate GARCH-MIDAS framework are the GARCH family models whose forecasting outcomes are examined. The outcome of GARCH-MIDAS analyses suggests that inflation, interest rate, exchange rate, and oil price are significant determinants of the volatility of the Johannesburg Stock Market All Share Index. While for Nigeria, the volatility reacts significantly to the exchange rate and oil price. Furthermore, inflation, exchange rate, interest rate, and oil price significantly influence Ghanaian equity volatility, especially for the long-term volatility component. The significant shock of the oil price and exchange rate to volatility is present in all three markets using the generalized autoregressive conditional heteroscedastic-mixed data sampling (GARCH-MIDAS) framework. The GARCH-MIDAS, with a powerful fusion of the GARCH model's volatility-capturing capabilities and the MIDAS approach's ability to handle mixed-frequency data, predicts the volatility for all variables better than the traditional GARCH framework. Incorporating these two techniques provides an innovative and comprehensive approach to modelling stock returns, making it an extremely useful tool for researchers, financial analysts, and investors.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 1997년도 추계학술대회발표논문집; 홍익대학교, 서울; 1 Nov. 1997
• /
• pp.105-108
• /
• 1997

Previous studies in stock market predictions using artificial intelligence techniques such as artificial neural networks and case-based reasoning, have focused mainly on spot market prediction. Korea launched trading in index futures market (KOSPI 200) on May 3, 1996, then more people became attracted to this market. Thus, this research intends to predict the daily up/down fluctuant direction of the price for KOSPI 200 index futures to meet this recent surge of interest. The forecasting methodologies employed in this research are the integration of genetic algorithm and artificial neural network (GAANN) and the integration of genetic algorithm and case-based reasoning (GACBR). Genetic algorithm was mainly used to select relevant input variables. This study adopts the categorical data preprocessing based on expert's knowledge as well as traditional data preprocessing. The experimental results of each forecasting method with each data preprocessing method are compared and statistically tested. Artificial neural network and case-based reasoning methods with best performance are integrated. Out-of-the Model Integration and In-Model Integration are presented as the integration methodology. The research outcomes are as follows; First, genetic algorithms are useful and effective method to select input variables for Al techniques. Second, the results of the experiment with categorical data preprocessing significantly outperform that with traditional data preprocessing in forecasting up/down fluctuant direction of index futures price. Third, the integration of genetic algorithm and case-based reasoning (GACBR) outperforms the integration of genetic algorithm and artificial neural network (GAANN). Forth, the integration of genetic algorithm, case-based reasoning and artificial neural network (GAANN-GACBR, GACBRNN and GANNCBR) provide worse results than GACBR.

• Journal of Preventive Medicine and Public Health
• /
• 제16권1호
• /
• pp.113-120
• /
• 1983

The main objective of this case study is to develop demand forecasting model for durg inventory control in a university hospital. This study is based on the pertinent records during the period of January 1975 to August 1981 in the pharmacy and stock departments of the hospital. Through the analysis of the above records the author made some major findings as follows: 1. In A.B.C. classification, the biggest demand (A class) consists of 9 items which include 6 items of antibiotics. 2. Demand forecasting level of an index or discrepancy in A class drug compared with real demand for 6 months is average 30.4% by X-11 Arima method and 84.6% by Winter's method respectively. 3. After the correcting ty the number of bed, demand forecasting of drug compared with real demand for 6 months is average 23.1% by X-11 Arima method and 46.6% by Winter's method respectively.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2000년도 추계학술대회 및 정기총회
• /
• pp.77-80
• /
• 2000

Feedback herding strategy in stock market means considering other investor's strategy as a basis of market forecasting of next term. Generally, individual investors use that strategy which mimics the strategy of institutional investors. When it is used in stock market, both kind of investors, preceders and followers, can take the higher average of rate of return to normal market in which no feedback herding strategy is not use, the more investors take part in. And variance of return, the risk of investment, are same to both group.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 1996년도 춘계공동학술대회논문집; 공군사관학교, 청주; 26-27 Apr. 1996
• /
• pp.745-748
• /
• 1996

Korea Composite Stock Price Index (KOSPI) was predicted using Cascade Correlation Network (CCN) model. CCN was suggested, by Fahlman and Lebiere [1990], to overcome the limitations of backpropagation algorithm such as step size problem and moving target problem. To test the applicability of CCN as a function approximator to the stock price movements, CCN was used as a tool for univariate time series analysis. The fitting and forecasting performance fo CCN on the KOSPI was compared with those of Multi-Layer Perceptron (MLP).

• 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.

• 전자공학회논문지CI
• /
• 제45권5호
• /
• pp.95-101
• /
• 2008

주가지수는 경제 및 정치적 상황을 포함한 다양한 주변 환경에 영향을 받는 관계로 정확한 주가지수 예측모형의 개발은 매우 어려운 문제로 여겨지고 있다. 본 논문에서는, 신경회로망과 통계적인 방법을 이용하여 종합주가지수(KOSPI)를 예측하는 에이전트 시스템 기법을 제안한다. 예측오차의 평균 및 편차를 최소화하기 위해서, 에이전트시스템은 특징추출, 변수선정, 예측 엔진선정 및 분석을 위한 부(sub)에이전트 모듈들을 포함하고 있다. KOSPI(Korea Composite Stock Price Index) 예측을 위한 에이전트시스템 구현의 첫 번째 단계로서, 주성분분석을 이용하여 22개의 표준기본경제지표에서 12개의 경제지표를 추출하였다. 열두 개의 추출된 경제지표들은 예측하고자하는 예측일에 따라 최량부분적합법을 이용하여 다시 한 번 입력 변수들을 선정하게 된다. 성능평가를 위해 주가지수의 변동폭이 다른 두 종류의 실험데이터를 대상으로 예측을 진행한 결과 30일의 연속적인 종합주가지수예측에 있어 11.92포인트의 평균오차율을 보였다. 또한, 예측시점에 따라 관련이 높은 기본지표의 종류 및 개수가 다르게 나타나므로 제안한 주가예측 에이전트시스템 구조가 유용함을 보였다.