인터넷 검색빈도와 아파트 거래량 간의 선행-후행 관계

유한수(Yoo, Han-Soo) 한국부동산정책학회 2020 不動産政策硏究 Vol.21 No.1

The aim of this paper is to investigate empirically the lead-lag relation between Internet search volume(Naver Trend Index) and trading volume of apartment. The Internet is the main source of information gathering. The increase of Internet search volume on something means the increase of people’s interest in something. Previous studies in this topic examine the relationship between Internet search volume and the released trading volume of apartment. The distinguishing feature of this empirical study is that it investigates the relation between Internet search volume and the fundamental trading volume of apartment, and the relationship between Internet search volume and the transitory trading volume of apartment. The first step of this empirical study is to decompose released trading volume into fundamental trading volume and transitory trading volume. The second step is unit root test. The third step is Granger causality test. The results of Granger causality test reveal that Internet search volume Granger causes released trading volume of apartment, and the fundamental trading volume of apartment. And there exists one-way Granger causality from Internet search volume to the transitory trading volume of apartment. The fourth step, the impulse response function analysis reveals that the shock of Internet search volume generally increases the magnitude of released trading volume, fundamental trading volume and transitory trading volume. The findings of this paper indicate that the movement of Internet search volume helps to estimate trading volume of apartment market. Therefore, the movement of Internet search volume can serve as an early market indicator.

한류가 우리나라 수출입에 미친 상호 영향에 관한 실증적 연구: BTS를 중심으로

임병진 한국무역연구원 2023 무역연구 Vol.19 No.1

Purpose – This study is an empirical study on the relationship among export volume and import volume, with BTS internet search trends indexed. Design/Methodology/Approach – In this study, we used 111 monthly data sets on BTS internet search trends, the export volume, and the import volume from April 30, 2013 to Jun 30, 2022. We try to analyze the mutual influence and causality among BTS internet search trends, the export volume, and the import volume. We seek to analyze the extent of any cross-influence. We employ a variance decomposition function based on a VAR model, as well as impulse response after a cointegration test and unit root test. Findings – The important results of this study are summarized as follows. First, there is at least one cointegration among the first differential data of BTS internet search trends, export volume, and import volume. Secondly, import volume does have Granger causality with BTS internet search trends. The export and import volume show Granger causality. BTS internet search trends and the export volume have Granger causality with BTS internet search trends and export volume. Research Implications – This study differs in that it uses BTS internet search trends and the Korean wave. It has limitations in that the study period is short and the study sample is limited.

인터넷 검색량과 투자자별 거래 및 주식수익률의 관계에 대한 실증 연구

김류미 ( Ryumi Kim ) 한국금융공학회 2018 금융공학연구 Vol.17 No.2

본 논문에서는 검색 엔진을 통해 기업을 검색한, 인터넷 검색량의 합계를 투자자 관심 척도로 하여, KOSPI시장에 상장된 모든 기업을 대상으로 투자자별 거래량과 주식수익률에 어떠한 영향을 미치는지 알아보았다. 인터넷 검색량은 국내 대표적인 검색 엔진인 네이버(NAVER)를 통해서 검색한 수치를 이용하였다. 기존 문헌과 마찬가지로 본 연구 표본에서도 인터넷 검색량은 독립적인 투자자관심변수로서 의미가 있었다. 결과에 따르면, 인터넷 검색량의 증가는 거래량의 증가를 가져오며, 모든 투자자의 매수와 매도 증가와 유의한 양의 관계가 있다. 특히 개인투자자의 매수 증가와의 양의 관계가 가장 강하다. 또한 인터넷 검색량이 많아지면, 주식수익률이 일시적으로 증가한다. 규모가 작은 기업일수록, 검색량 증가에 따른 주식수익률 증가가 더 크다. 기존 문헌들을 참고하면, 이러한 검색량과 주식수익률 간의 관계는 개인투자자들의 일시적인 매수압력에 의한 것으로 판단된다. 따라서 매수압력으로 인해 주식수익률의 영향을 많이 받는 작은 기업에서 이러한 양의 관계가 더 크게 되는 것이다. 마지막으로, 본 연구 분석에서는 개인투자자의 거래회전율이 높은 주식, 그리고 개인투자자의 순매수가 높은 주식에서, 검색량이 높을수록 주식수익률이 더 높았다는 것을 직접적으로 증명하였다. I use aggregate search volume in internet portal site as a measure of investor attention to securities. and investigate the relationship between investor attention, investors trading, and stock returns in KOSPI market. According to the results, search volume can be the independent investor attention proxy. An increase in aggregate search volume leads an increase in trading volume, and is positively correlated with an increase in both buying and selling of all investors. In especial, an increase in individual investors' buying has the strongest positive correlation with the increase in search volume. This finding supports that search volume captures individual investor attention. Further, stocks with increased search volume have temporary higher returns. This relation is stronger among small stocks. Literature argues that the positive relation between asset price and search volume is caused by price pressure due to individual buying activity. I directly find that increased search volume leads an increase in individual investors’ net buying, and has stronger positive relation with returns on stocks with higher individual investors’ turnover and net buying.

인공지능의 사회적 수용도에 따른 키워드 검색량 기반 주가예측모형 비교연구

조유정(Yujung Cho),손권상(Kwonsang Sohn),권오병(Ohbyung Kwon) 한국지능정보시스템학회 2021 지능정보연구 Vol.27 No.1

Recently, investors interest and the influence of stock-related information dissemination are being considered as significant factors that explain stock returns and volume. Besides, companies that develop, distribute, or utilize innovative new technologies such as artificial intelligence have a problem that it is difficult to accurately predict a companys future stock returns and volatility due to macro-environment and market uncertainty. Market uncertainty is recognized as an obstacle to the activation and spread of artificial intelligence technology, so research is needed to mitigate this. Hence, the purpose of this study is to propose a machine learning model that predicts the volatility of a companys stock price by using the internet search volume of artificial intelligence-related technology keywords as a measure of the interest of investors. To this end, for predicting the stock market, we using the VAR(Vector Auto Regression) and deep neural network LSTM (Long Short-Term Memory). And the stock price prediction performance using keyword search volume is compared according to the technologys social acceptance stage. In addition, we also conduct the analysis of sub-technology of artificial intelligence technology to examine the change in the search volume of detailed technology keywords according to the technology acceptance stage and the effect of interest in specific technology on the stock market forecast. To this end, in this study, the words artificial intelligence, deep learning, machine learning were selected as keywords. Next, we investigated how many keywords each week appeared in online documents for five years from January 1, 2015, to December 31, 2019. The stock price and transaction volume data of KOSDAQ listed companies were also collected and used for analysis. As a result, we found that the keyword search volume for artificial intelligence technology increased as the social acceptance of artificial intelligence technology increased. In particular, starting from AlphaGo Shock, the keyword search volume for artificial intelligence itself and detailed technologies such as machine learning and deep learning appeared to increase. Also, the keyword search volume for artificial intelligence technology increases as the social acceptance stage progresses. It showed high accuracy, and it was confirmed that the acceptance stages showing the best prediction performance were different for each keyword. As a result of stock price prediction based on keyword search volume for each social acceptance stage of artificial intelligence technologies classified in this study, the awareness stages prediction accuracy was found to be the highest. The prediction accuracy was different according to the keywords used in the stock price prediction model for each social acceptance stage. Therefore, when constructing a stock price prediction model using technology keywords, it is necessary to consider social acceptance of the technology and sub-technology classification. The results of this study provide the following implications. First, to predict the return on investment for companies based on innovative technology, it is most important to capture the recognition stage in which public interest rapidly increases in social acceptance of the technology. Second, the change in keyword search volume and the accuracy of the prediction model varies according to the social acceptance of technology should be considered in developing a Decision Support System for investment such as the big data-based Robo-advisor recently introduced by the financial sector.

A Study on the Guarantee Supply of Regional Credit Guarantee Foundations Using Hierarchical Regression: Using Naver Trend & Mediation Effect

Huifeng Pan,Hee-Young Son,Man-Su Kang 보안공학연구지원센터 2016 International Journal of Software Engineering and Vol.10 No.4

The data used in this study includes guarantee-related data from January 2007 to September 2014, as well as data of search volume provided by Naver Trend. As a result of the hierarchical regression analysis, it turned out that in step 1, the volume of credit guarantee searching affected guarantee supply while the default amount did not. In addition, it is expected that as the volume of credit guarantee searching increases, guarantee supply will also be expended. In step 2, it turned out that the amount of default normalization affected guarantee supply. It is expected that as the amount of default normalization increases, guarantee supply will also be expanded. Finally, in step 3, it turned out that both the volume of credit guarantee searching and the amount of default normalization affected guarantee supply while the default amount did not. As for the explanation power of the three models, that of step 1 was 8.3%, that of step 2 was 13.7%, and that of step 3 was 19.3%. As the default amount affected guarantee supply, Sobel Test was conducted to measure the mediation effect of the volume of credit guarantee searching and the amount of default normalization. As a result, it turned out that the volume of credit guarantee searching had mediation effect while the amount of default normalization did not. The objective of this study is to examine the effect of guarantee accidents of the Regional Credit Guarantee Foundation, which provides public finance service on guarantee supply. This indicates the necessity of examining the causality through another regression analysis after a Granger Causality test. In terms of demands for guarantee supply, further research may be necessary on referring to the volume of credit guarantee searching.

소비자의 인터넷 검색활동이 농축산물 구매에 미치는 영향

노호영,김성용,김태영 한국농업경제학회 2019 農業經濟硏究 Vol.60 No.2

This paper analyzed how the internet search volume provided by Naver Data Lab, the largest search engine in Korea, affects consumers purchases of agricultural food products. The subjects of this study were beef, pork, chicken, and chinese cabbage, radish, red pepper, garlic, and onion with the markets being very vulnerable in terms of the supply and demand. Using the consumers panel data of 1,314 households in the year of 2016-2017 provided by the Rural Development Administration, we analyzed whether the internet search volume for each product had a significant effect on consumers purchases of the food product, and estimated the marginal effect of its internet search volume. The main results of this paper are as follows. First, internet search activities on beef, chicken, cabbage, radish, red pepper, garlic and onion were found to be the factor that increased consumers purchases of the products. When the internet search volume for each food product was increased by 1 unit, the purchase amount of the product was increased by about 1% of its average purchase amount. In this context, the internet search volume can become one of the effective tools to quickly and easily identify the psychology or behaviors of those who purchase agricultural food products. Therefore, data obtained from consumers internet search activities could be employed to provide useful information in terms of making marketing strategies of agricultural producers and food companies, and establish more quickly and appropriately policy options in response to changes in food demand and supply.

인터넷 검색 활동과 주택 가격 및 거래량 간 동적 관계 분석

김대원,유정석 한국부동산연구원 2014 부동산연구 Vol.24 No.2

주택 시장에서 인터넷 검색 활동의 흔적은 집단적 사고이자 장래 주택 구매 수요의 반영을 의미한다. 따라서 인터넷 검색 활동과 주택 가격 및 거래량 간 유의미한 관계를 포착할 수 있다면, 이는 곧 인터넷 검색 활동이 주택 구매 수요의 대리 변수로서 미래 주택 시장을 예측할 수 있는 지표로 활용될 수 있음을 의미하게 된다. 이러한 맥락에서 본 연구는 인터넷 검색 활동과 주택 가격 및 거래량 간의 동적 관계를 확인하기 위하여 서울시 23개 행정구, 2007년 1월부터 2014년 2월까지를 연구 대상 및 범위로 설정하고, 인터넷 검색 활동의 대리 변수로서 “네이버 트렌드” 자료를, 주택 가격 및 거래량의 대리 변수로는 아파트매매가 지수 및 아파트매매거래량 지수 패널 자료를 사용하여 실증 분석을 실시하였다. Arellano-Bond 동적 패널 모형 추정 결과, 인터넷 검색 활동은 일정 기간의 시차를 두고 주택 가격 및 거래량에 유의미한 양(+)의 영향을 미치는 것으로 확인되었다. 패널 VAR 모형을 통한 IRFs 및 FEDVs 분석 결과, 인터넷 검색 활동은 주택 가격 및 거래량에 1차 시차에서 가장 큰 양(+)의 영향을 미치며, 주택 거래량 보다는 주택 가격에 더 큰 비중으로 영향을 미침을 확인할 수 있었다. 패널 Granger 인과성 검정 결과, 인터넷 검색 활동과 주택가격, 주택 가격과 거래량은 상호 순환적 인과성을 갖는 것으로 나타났다. In the housing market, the internet searching activity trail means collective thinking and represents purchase intent. Therefore, if we can capture the meaningful relationship between the internet searching activity, the housing price, and the housing trading volume, it would mean we can predict the future using the internet searching activity data as a index for the housing market. In this context, we conducted the empirical research to examine the dynamic relationship between the internet searching activity and the housing price and trading volume. Using the “NAVER Trend" data as a proxy for the internet searching activity, the apartment sale price index as a proxy for the housing price, and the apartment trading volume index as a proxy for the housing trading volume, we set up the panel data of 23 autonomous districts in Seoul form Jan. 2007 to Feb. 2014. In results from the Arellano-Bond dynamic panel model, we found that the internet searching activity had an positive(+) effect on the housing price and trading volume with some time lags. From the results of IRFs and FEVDs derived from the panel VAR model analysis, we also found that the internet searching activity had the strongest effect on housing market at the first order lag and influences much more on the housing price rather than the trading volume. In addition, panel Granger causality test results showed that the internet searching activity and the housing price, the housing price and trading volume had mutual cyclic causalities each other.

주식시장의 빅데이터 검색트렌드와 KOSPI 및 KOSPI 거래량의 상호 작용에 관한 연구

임병진 국제e-비즈니스학회 2023 e-비즈니스 연구 Vol.24 No.4

Research Purpose: This study investigates the effects among the KOSPI search index, KOSPI index and trading volume in Korea. Research Methods: We employ variance decomposition function for the KOSPI search index, KOSPI index, and trading volume based on VAR model as well as impulse response. Results in Research: An important result of this study are summarized as follows: First, as a result of the stability test for the level variable of the KOSPI search index, KOSPI index and trading volume, it was found to be unstable at the 5% significance level. Second, it was found that the results of the stability test for the first difference data of the KOSPI search index, KOSPI index and trading volume were all stable at the 1% significance level. Research Conclusion: Mutual influence was observed between the KOSPI index and the KOSPI search index. The KOSPI search index was found to be a causal variable affecting the KOSPI trading volume, and the KOSPI index was found to be a causal variable affecting the KOSPI trading volume. 이 연구는 KOSPI 검색지수와 KOSPI 지수 및 거래량 간의 상호 미치는 영향을 실증적으로 분석한연구이다. KOSPI 검색지수와 KOSPI 및 거래량의 일간 시계열 자료는 2020년 1월 2일부터 2022년 11월18일까지 713개의 일간 자료이다. KOSPI 검색지수와 KOSPI 및 거래량의 일간 시계열 자료간의 인과관계와상호영향력을 살펴봄으로써 KOSPI 검색지수와 KOSPI 및 거래량의 일간 시계열 자료 간의 미치는 영향력의정도를 분석하고자 한다. KOSPI 검색지수와 KOSPI 및 거래량의 일간 시계열 자료의 단위근 검정, VAR 모형 분석을 이용하여 KOSPI 검색지수와 KOSPI 및 거래량의 일간 시계열 자료의 예측오차 분산분해와충격반응분석, KOSPI 검색지수와 KOSPI 및 거래량의 일간 시계열 자료 간의 Granger인과관계 검정을 실시하였다. KOSPI 검색지수와 KOSPI 지수 및 거래량에 관한 연구의 결과는 다음과 같다. 첫째, KOSPI 거래량일간 자료는 1% 유의수준에서 수준변수의 안정성검정 결과는 불안정적으로 나타났다. 둘째, KOSPI 검색지수와 KOSPI 지수 및 거래량 일간 자료의 제1차 차분 변수에 안정성검정 결과는 1% 유의수준에서 모두 안정적임을 알 수 있었다. 셋째, KOSPI 거래량 분산분해에서 KOSPI 거래량의 변화는 KOSPI 거래량 자체의 내재적변화가 97% 이상을 설명하고 있고, KOSPI 검색지수의 설명력은 3% 이상을 설명하고 있고 KOSPI 지수의설명력은 4% 이상을 설명하고 있다. 넷째, KOSPI는 KOSPI 검색지수는 상호 간에 각각 상호 영향을 미치는것으로 나타났고 KOSPI 검색지수는 KOSPI 거래량에 영향을 원인변수로 나타났고 KOSPI은 KOSPI 거래량에영향을 미치는 원인변수인 것으로 나타났다. 마지막으로 KOSPI 검색지수와 KOSPI 간의 상관계수는-0.536824로 음(-)의 관계를 나타내고 KOSPI 검색지수와 KOSPI 거래량 간의 상관계수는–0.214664로 음(-)의관계를 보여주고 있고 KOSPI 거래량과 KOSPI 간의 상관계수는 0.005029로 양(+)의 관계를 나타내고 있다.

뉴스 빅 데이터와 구글 정보 검색이 아파트 거래량에 미치는 영향에 관한 실증 연구

김윤식(Kim, Yun Sik),최민섭(Choi, Min Seub) 한국주거환경학회 2018 주거환경(한국주거환경학회논문집) Vol.16 No.2

The purpose of this study is to investigate the relationship between news big data and google information search and to analyze the cause of apartment transaction volume. The effect of this on the apartment trading volume was examined. Moreover, this study examined the relation between real estate information search and news big data, and tried to analyze what kind of causal relationship exists in apartment transaction volume. And the demonstrate and study the impact on the apartment transaction volume. The results of the study are as follows: First, it was found that the increase in the number of news big data articles affected the amount of apartment transactions. Second, it was found that the increase in Google search affects apartment volume, there is mutual causal relationship. Last, As the news big data and Google search increases the apartment transaction price index did not increase.

빅데이터를 통해 본 한국 현대 시인 인지도 및 평판분석

최도식 이화여자대학교 이화인문과학원 2018 탈경계인문학 Vol.11 No.2

Which of the modern poets of Korea will be frequently found in the public? Which poems will be found in the public? To answer this question, they used Big Data, which has recently been in the spotlight. The study's data collection and analysis used Google Trends, thumb trends, and social metrics. The analysis procedure searched for 100 poets with Google Trends. 27 people with high frequency of search were compared. The study had a reputation analysis of 13 of the 27 poets. The first analysis of the study showed that Kim So-wol, Han Yong-woon, Seo Jung-joo, Kim Chun-soo and Jung Ji-yong had a high number of searches. The relevant search result for the study was Kim So-wol's "Azaleas", "Invocation of the spirits of the dead", "Some distant day", "A mountain flower", "My house" and Han Yong-woon's "Your silence", "I don't know", "Obedience", Seo Jeong-joo's "By Chrysanthemum", "Self-portrait", Kim Chun-Soo is "Flower", Jung Ji-young is "Fragrance", "Lake", "Springtime". Yoon Dong-joo is "Prologue poem", "Starry Night", "Self-portrait", "Cross", "Confession", "Hometown", "Written easily", "Shoot the Moon". The Poet Kim Soo-young had many searches for the comedian Kim Soo-young. The living poets are the order of An Do-hyeun, Do Jong-hwan, Jung Ho-seung, Kim Ji-ha, and Kim Yong-taek. The search word of An Do-hyeon discovered as "Ask you", "Permeating there", "A piece of briquette" and 'a used briquet', 'briquette' of poetic diction. Do Jong-hwan discovered as "You, the Hollyhock", "An ivy" and 'senator', 'minister', poet Jung Ho-seung discovered as "Spring road", "To Daffodil", "Unbearable letter". And Kim Ji-ha searched out "Burning thirst", Kim Yong-taek searched out "Seomjin River", "You are so good", "Spring day" and "A wild chrysanthemum". However, the poet Ko eun was more searched by Ko eun, the leader of the girl group. In the analysis of Opinion Mining, affirmative responses of related words are in order of Do Jong-hwan, Kim Yong-taek, Kim Chun-Soo, Han Yong-woon, Jung Ho-seung, An Do-hyeon, Jung Ji-young, Kim So-wol, Seo Jeong-joo, Yoon Dong-joo, Kim Ji-ha. Yoon Dong-joo's neutral reaction was interpreted as disturbance factor related to internet shopping mall. Yoon Dong-joo's search volume in Google Trent was excessive due to sales of Internet shopping malls. In the analysis of Opinion Mining, the highest order of positive response was analyzed by Kim So-wol, Han Yong-woon, Yoon Dong-joo, Jung Ji-young, Do Jong-hwan, Kim Yong-taek, Jung Ho-seung, An Do-hyeon. Ko eun poet and Kim Ji-ha poet have received increasingly negative reputation from the public. An Do-hyeon, Jung Ho-seung, Do Jong-hwan and Kim Yong-taek are expected to increase their positive reputation gradually. Therefore, the translation work for the globalization of modern Korean poetry should be focused on the poems of An Do-hyeon, Jung Ho-seung, Do Jong-hwan and Kim Yong-taek.