• Journal of the Korea Society of Computer and Information
• /
• v.27 no.8
• /
• pp.31-39
• /
• 2022

Stock prices are influenced by a number of external factors, such as laws and trends, as well as number-based internal factors such as trading volume and closing prices. Since many factors affect stock prices, it is very difficult to accurately predict stock prices using only fragmentary stock data. In particular, since the value of a company is greatly affected by the perception of people who actually trade stocks, emotional information about a specific company is considered an important factor. In this paper, we propose a deep learning-based stock price prediction model using sentiment analysis with news data considering temporal characteristics. Stock and news data, two heterogeneous data with different characteristics, are integrated according to time scale and used as input to the model, and the effect of time scale and sentiment index on stock price prediction is finally compared and analyzed. Also, we verify that the accuracy of the proposed model is improved through comparative experiments with existing models.

• Journal of Cadastre & Land InformatiX
• /
• v.50 no.2
• /
• pp.133-153
• /
• 2020

Officially assessed land price has been the index of South Korea since 1989 throughout different sectors of tax and welfare. Officially assessed land price is used as a tax valuation for the tax on property holdings, and the equity of such is the most important factor in the fair taxation for the people of South Korea. On this wise, this research analyzed and verified the horizontal and vertical inequity of officially assessed land price in Seoul by using the real transaction data between 2016 and 2018. In fact, Seoul's assessment ratio for the entire three-year period was 60.64% and it showed to increase each year. Horizontal equity was found to be most favorable in 2017, and the horizontal equity of each borough of Seoul appeared to improve each year. Vertical inequity was found to have reverse inequality in most boroughs of Seoul, however, some parts of Gangnam districts such as Gangnam-gu, Seocho-gu, and Gangdong-gu presented progressive inequality. Such example showed the need for improvement in terms of balance by each borough. The use of quantile regression demonstrated reverse inequality in most quantile, but, the differences in the value of the coefficient by each quantile showed the need for improvement of officially assessed land price with the equity of each quantile. Through the equity verification of officially assessed land price, it was analyzed that the lack of equity was found by year, by borough, and by use district. In order to redeem the lack of equity, the government must systematically supplement the real-estate disclosure system by initiating ratio studies to verify horizontal and vertical equity.

• The Korean Journal of Financial Management
• /
• v.12 no.2
• /
• pp.239-271
• /
• 1995

본 연구는 현재 미국에서 거래되고 있는 세 가지 주가지수선물 상호간의 일중(intradaily) 가격선도(price leadership) 관계에 관한 실증분석이다. 본 연구가 기존의 연구와 다른점은, 기존의 연구가 주가지수선물과 그 기준이 되는 현물 가격사이의 가격 선도 관계에 초점을 두고 있는데 반하여 본 연구는 주가지수선물 시장 사이에서 존재하는 가격 선도관계를 분석하고 있다는 점이다. 실증 분석의 대상이 된 주가지수선물들은 Chicago Mercantile Exchange의 Standard and Poor's 500 Index(S&P 500), New York Futures Exchange의 New York Stock Exchange Composit Index (NYSE), 그리고 Chicago Board of Trade의 Major Market Index(MMI)이다. 만약 이들 시장들이 정보의 전달에 있어서 효율적(informationally efficient) 이라면 이들 가격간에 선도-지연(lead-lag) 현상은 존재하지 않을 것이다. 그러나 어느 한 시장이 새로운 정보를 선물가격에 반영하는데 다른 시장에 비해 상대적으로 느리다면, 이들 시장 상호간에는 가격의 전이(transmission)현상이 존재하게 될 것이다. 이들 선물간의 일중 가격선도 관계 연구는 이러한 시장의 효율성 문제를 밝히는데 의의가 있을 뿐만 아니라, 시장간의 단기적 가격 괴리를 이용하려는 차익거래자들에게도 유용하게 쓰일 수 있을 것이다. 본 연구는 위에서 언급한 각각의 주가지수선물들이 가격 선도성을 가질 수 있는 이유와 관련된 다음과 같은 세 가지 가설을 설정하였다. 첫째 가설은, 가격의 선도성은 거래량과 관련이 있다는 것이다. 즉, 이들 주가지수선물 중 가장 거래량이 많은 S&P 500 선물이 다른 선물을 선도할 것이라는 가설이다. 둘째, 가격의 선도성은 주가지수를 구성하는 주식의 수에 비례한다는 가설이다. 다시 말하면, 보다 않은 수로 구성된 주가지수일수록 정보처리 속도가 빠르다는 가설이다. 따라서, 본 연구에 포함된 주가지수선물 중 가장 많은 수의 주식을 대상으로 하는 NYSE 선물이 다른 선물을 선도할 것이다. 마지막 가설은 정보의 처리는 대형주 혹은 기관선호주(institutionally-favored)들이 주도한다는 것이다. 따라서, 주로 이와 같은 주식들로 구성 된 MMI 선물이 선도성을 가질 수 있다는 것이다. 위의 가설들을 검증하고 시장간의 가격 선도관계를 분석하기 위하여 본 연구는 vector autoregressive(VAR) 모형을 이용하여 충격-반응 함수(impulse response functions)를 계산하고, 분산분해(variance decomposition)를 수행하였다. 또한 가격 상호간에 존재할지도 모르는 공적분(cointegration)관계를 Johansen(1991)과 Jokansen and Juselius (1992) 등이 제시한 다변량 공적분 검정(multivariate cointegration test)를 통하여 분석하였다. 분석기간은 1986년 1월부터 1990년 7월까지이며, 각 주가지수선물들의 5분 간격 data를 사용하였다. 연구결과, 충격-반응 분석은 어느 한 시장에서의 충격(shock)은 다른 시장으로 매우 빠르게 전달되고 있음을 보여 주었다. 그러나 충격의 지속정도는 그 충격의 진원지에 따라 달랐다. 즉, NYSE나 MMI 선물로부터 발생 한 충격은 다른 시장의 가격에 5분 안에 반영을 끝냈지 만 S&P 500 선물에서 발생한shock은 그 이상 지속되었다. 또한, 분산분해 결과 S&P 500 선물이 자기자신 뿐만 아니라 다른 시장의 예상하지 못했던 움직임(unexpected movements)을 설명하는데 가장 큰 설명력(explanatory power)을 가지고 있었다. 결론적으로 S&P 500 선물이 다른 선물을 약 5분 간격으로 선도하였다. 이는 가격의 선도가 거래량과 밀접한 관계가 있음을 보여 주는 것이다.

• Proceedings of the Korean Institute Of Construction Engineering and Management
• /
• autumn
• /
• pp.423-426
• /
• 2002

The foundation for developing a cost estimation system based on historical data has been being prepared in Korea. Historical data is a priori of developing a cost estimation model. Cost Index, one of the historical data, is used to estimate construction cost and to adjust the amount of contract money in the foreign country, whereas it is not used in domestic except for the road construction project in Korea. Construction cost indices can be used by an estimator in tender analysis, pricing, price adjustment, cost planning, and forecasting. In this regards, this paper identified the problems in developing Cost Index evaluation process by comparing the standard of framing Cost Index used in British to the one used in Korea. Then, the scheme for improving a Cost Index required for Site Developing Construction was proposed. Twenty-two cases of engineering estimate data were used to compare the domestic standard to the foreign one in deriving a Cost Index.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.1D
• /
• pp.105-113
• /
• 2006

Construction cost index is generally used to estimate the new project cost based on past construction data and to adjust the contract cost when the price change of various articles and items of expenditure composing the contract occurs. In Korea, it is mostly used for modulation of construction contract cost due to fluctuation of prices. However the method for making cost index had some problems in calculating cost index of each expenditure item that could not properly reflect the change of construction cost. To supplement these problems, the research of developing construction cost index has been executed. Through the precedent research, these problems were partially resolved but still remain. Therefore this research proposes the method for making cost index that utilizes representative items of labor, material, equipment by analyzing bill of quantity of road construction, through analysis and comparison of precedent studies. By using this method, it is expected to solve the problems which were not reflected in preceeding studies.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.7
• /
• pp.73-80
• /
• 2024

In this paper, we proposes a method to improve the accuracy of predicting the Korea Composite Stock Price Index (KOSPI) by combining topic modeling and Long Short-Term Memory (LSTM) neural networks. In this paper, we use the Latent Dirichlet Allocation (LDA) technique to extract ten major topics related to interest rate increases and decreases from financial news data. The extracted topics, along with historical KOSPI index data, are input into an LSTM model to predict the KOSPI index. The proposed model has the characteristic of predicting the KOSPI index by combining the time series prediction method by inputting the historical KOSPI index into the LSTM model and the topic modeling method by inputting news data. To verify the performance of the proposed model, this paper designs four models (LSTM_K model, LSTM_KNS model, LDA_K model, LDA_KNS model) based on the types of input data for the LSTM and presents the predictive performance of each model. The comparison of prediction performance results shows that the LSTM model (LDA_K model), which uses financial news topic data and historical KOSPI index data as inputs, recorded the lowest RMSE (Root Mean Square Error), demonstrating the best predictive performance.

• Journal of Intelligence and Information Systems
• /
• v.22 no.2
• /
• pp.81-96
• /
• 2016

As the internet has become widespread and easy to access everywhere, it is common for people to search information via online search engines such as Google and Naver in everyday life. Recent studies have used online search volume of specific keyword as a measure of the internet users' attention in order to predict disease outbreaks such as flu and cancer, an unemployment rate, and an index of a nation's economic condition, and etc. For stock traders, web search is also one of major information resources to obtain data about individual stock items. Therefore, search volume of a stock item can reflect the amount of investors' attention on it. The investor attention has been regarded as a crucial factor influencing on stock price but it has been measured by indirect proxies such as market capitalization, trading volume, advertising expense, and etc. It has been theoretically and empirically proved that an increase of investors' attention on a stock item brings temporary increase of the stock price and the price recovers in the long run. Recent development of internet environment enables to measure the investor attention directly by the internet search volume of individual stock item, which has been used to show the attention-induced price pressure. Previous studies focus mainly on Dow Jones and NASDAQ market in the United States. In this paper, we investigate the relationship between the individual investors' attention measured by the internet search volumes and stock price changes of individual stock items in the KOSDAQ market in Korea, where the proportion of the trades by individual investors are about 90% of the total. In addition, we examine the difference between industries in the influence of investors' attention on stock return. The internet search volume of stocks were gathered from "Naver Trend" service weekly between January 2007 and June 2015. The regression model with the error term with AR(1) covariance structure is used to analyze the data since the weekly prices in a stock item are systematically correlated. The market capitalization, trading volume, the increment of trading volume, and the month in which each trade occurs are included in the model as control variables. The fitted model shows that an abnormal increase of search volume of a stock item has a positive influence on the stock return and the amount of the influence varies among the industry. The stock items in IT software, construction, and distribution industries have shown to be more influenced by the abnormally large internet search volume than the average across the industries. On the other hand, the stock items in IT hardware, manufacturing, entertainment, finance, and communication industries are less influenced by the abnormal search volume than the average. In order to verify price pressure caused by investors' attention in KOSDAQ, the stock return of the current week is modelled using the abnormal search volume observed one to four weeks ahead. On average, the abnormally large increment of the search volume increased the stock return of the current week and one week later, and it decreased the stock return in two and three weeks later. There is no significant relationship with the stock return after 4 weeks. This relationship differs among the industries. An abnormal search volume brings particularly severe price reversal on the stocks in the IT software industry, which are often to be targets of irrational investments by individual investors. An abnormal search volume caused less severe price reversal on the stocks in the manufacturing and IT hardware industries than on average across the industries. The price reversal was not observed in the communication, finance, entertainment, and transportation industries, which are known to be influenced largely by macro-economic factors such as oil price and currency exchange rate. The result of this study can be utilized to construct an intelligent trading system based on the big data gathered from web search engines, social network services, and internet communities. Particularly, the difference of price reversal effect between industries may provide useful information to make a portfolio and build an investment strategy.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1998.10a
• /
• pp.224-227
• /
• 1998

This paper proposes inter-level causal reasoning to implement synergistic approach. We decompose KOSPI prediction model into economy and industry level. Two kinds of intra-level QCOM are combined in inter-level QCOM via Inter-level relations. Downward reasoning is achieved by propagating the disturbance in the higher level to lower level while upward reasoning is to analyze the reverse cases.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1998.10a
• /
• pp.228-231
• /
• 1998

Artificial Intelligence literatures have recognized that stock market is a highly unstructured and complex domain so that it is difficult to find knowledge that belongs to that domain. This paper demonstrates that the proposed QCOM can derive global knowledge about stock market on the basis of a set of local knowledge and express it as a digraph representation. In addition, inference mechanism using quantitative causal reasoning can describe the qualitative and quantitative effects of exogenous variables on stock market.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10b
• /
• pp.320-322
• /
• 2000

본 논문에서 제안한 종합주가지수 "왼쪽어깨" 패턴 검출은 자기 연상 학습 신경망을 사용하였다. 종합주가 지수 데이터에서 머리어깨모형 중 왼쪽 어깨에 해당하는 데이터로 신경망을 학습시킨 후 이를 이용하여 현재 혹은 테스트 데이터를 입력으로 주어 성능을 평가하였다. 결과는 비교적 우수하였다. 패턴 검출에 의한 투자를 하였을 경우 17개월간의 누적 수익률이 132% 였다. 이 기간동안 buy and hold 전략을 사용했을 경우의 수익률은 39% 였다.률은 39% 였다.