유가충격이 한국의 주식시장에 미치는 영향: Granger 및 Toda-Yamamoto 인과성검정을 중심으로

조하현 ( Ha Hyun Jo ),김재혁 ( Jae Hyeok Kim ) 연세대학교 경제연구소 2015 延世經濟硏究 Vol.22 No.1

본 연구에서는 유가충격이 한국의 주가지수에 미치는 영향을 분석하였다. 국제유가, 산업생산지수, 코스피지수의 3변량 VECM과 VAR(p+d) 를 추정하였고 추정치를 이용하여 그랜져 인과성(Granger Causality) 과 Toda-Yamamoto 인과성을 검정하였다. 그 결과 국제유가는 코스피지수를 5% 유의수준에서 단기 그랜져 인과(Granger Cause)하였으며10% 유의수준에서 장기 그랜져 인과(Granger Cause)하였다. 따라서10% 유의수준에서 국제유가는 코스피지수를 강 인과(Strong Cause) 함을 확인하였다. Toda-Yamamoto 인과성 검정에서도 국제유가가 코스피지수를 장기적으로 인과(Cause)한다는 동일한 결과를 얻을 수 있었다. 또한, 국제유가의 상승충격은 우리나라의 주가지수를 하락시키는 것으로 나타났다. In this paper, we investigate the effect of crude oil price shocks on the Korean KOSPI. We estimate the VECM and VAR(p+d) using the three variables, such as crude oil price, industrial production and KOSPI index. We further analyze Granger Causality and Toda-Yamamoto Causality. The results of Causality tests show that the crude oil price Granger Cause KOSPI at 5% significance level in the short-run and at 10% significance level in the long-run. Hence, the crude oil price strongly Granger Cause KOSPI at 10% significance level. In Toda-Yamamoto Causality, we obtain the same result that the crude oil price cause KOSPI in the long-run. Additionally, we find that the positive crude oil price shock lowers KOSPI.

한국 주가지수 변동성과 일본 주가지수 변동성

유한수 ( Han Soo Yoo ) 아시아.유럽미래학회 2010 유라시아연구 Vol.7 No.1

In this paper, I investigate empirically lead-lag relationship between KOSPI200 volatility and NIKKEI225 volatility. There is a strong economic relationship between Korea and Japan in investment and international trade. Previous studies have been concentrated on the first moment of stock price index and observed volatility. In research angle, the analysis in this study is distinguished from previous studies in that this paper is concentrated on permanent volatility and transitory volatility. That is, the contribution of this paper is investigating the relationship between KOSPI200 permanent volatility and NIKKEI225 permanent volatility and the relationship between KOSPI200 transitory volatility and NIKKEI225 transitory volatility. Permanent volatility means the long-run component of observed volatility and transitory volatility means the short-run component of observed volatility. Observed volatility, permanent volatility and transitory volatility are estimated by using Component GARCH model. In Component GARCH model, observed volatility is assumed as the sum of permanent component and transitory component. Permanent volatility component has nearly a unit root, and transitory volatility component has a rapid time decay. That is, the long-run component is more persistent and mean-reverts much slower than short-run component. The sample period in this paper ranges from 4 January 2005 to 30 October 2009. Because Korean stock market holidays differ from Japanese stock market holidays, the corresponding date in another country which is a holiday in one country is deleted in the data set. I check the statistical features of the data. The summary statistics of stock price index return are as follows. KOSPI200 and NIKKEI225 time series data have the features of stock return as spiked peak and persistence. Ex ante ARCH LM tests show that there are ARCH effects in the residuals. Therefore, GARCH type is appropriate for these series. The procedure used in this paper involves the following steps. First, to estimate observed volatility, permanent volatility and transitory volatility, this study employs Component GARCH model. In Component GARCH model, variance equation is expressed as follows: σ2t = qt+a1(ε2t-1-qt-1)+β1(σ2t1-qt-1), qt=ω+p(qt-1-ω)+φ(ε2t-1-σ2t-1). That is, observed volatility is assumed as the sum of permanent volatility component and transitory volatility component. The Component GARCH model estimation results show that AR coefficients of long-run component of KOSPI200 and NIKKEI225, p, are nearly one, and persistent rates of short-run component, a1+β1, are smaller than p. In the second step, before examining the lead-lag relations, the unit root test to check the stationarity of each series is first employed. To test for the existence of a unit root, Augmented Dickey-Fuller test is used. The appropriate lag length is determined by the Schwartz Criterion. In each case of volatilities, ADF test results show that the null hypothesis that the series has a unit root is rejected. That is, observed volatility, permanent volatility and transitory volatility are stationary series. Therefore, level series is used in VAR model for Granger causality test in this study. In the third step, to examine the lead-lag relationship between KOSPI200 volatility and NIKKEI225 volatility, Granger causality tests are performed. Granger causality test is a useful tool to uncover the direction of causality. Granger causality test examines whether the past values of a variable, NIKKEI225 volatility, helps explain the current value of another variable, KOSPI200 volatility. Observed volatility, permanent volatility and transitory volatility data are found to be stationary series. Therefore, Granger causality test based on VAR is applied. The Granger causality tests show that there are bidirectional Granger causalities between KOSPI200 observed volatility and NIKKEI225 observed volatility. Permanent volatility is the trend component of observed volatility, that is, it is the long-run component of observed volatility. As expected, there are bidirectional Granger causalities between KOSPI200 permanent volatility and NIKKEI225 permanent volatility. Transitory volatility is the cyclical component of observed volatility, that is, it is the short-run component. The Granger causality test based on VAR shows that there are bidirectional Granger causalities between KOSPI200 transitory volatility and NIKKEI225 transitory volatility. In short, the finding of this study suggests that NIKKEI225 volatilities help to estimate KOSPI200 volatilities. Therefore, analyzing the NIKKEI225 volatilities is required for managing KOSPI200 related financial instruments and implementing global hedging strategies and capital asset pricing. In this paper, Component GARCH model is employed to estimate permanent volatility and transitory volatility. To compare my findings, the area for further work is to do more applied work by using other statistical methods estimating permanent volatility and transitory volatility.

국내 제조업부문의 에너지소비, 생산, 수출간의 인과관계 분석

김수이 ( Suyi Kim ) 한국환경경제학회 한국자원경제학회 2017 자원·환경경제연구 Vol.26 No.1

본 연구에서는 우리나라의 제조업을 대상으로 에너지소비, 생산, 수출의 상호 인과관계를 분석하였다. 우리나라 제조업을 9개 산업으로 나누어 1991년부터 2013년까지 패널 데이터를 구축하여 VECM 방법론과 더불어 Demitrescu and Hurlin (2012)에 의해서 개발된 패널Granger causality test 방법을 사용하였다. 분석결과에 의하면, 생산에서 에너지소비로, 수출에서 에너지소비로의 Granger Causality가 존재하였다. 하지만 그 역으로는 Granger Causality가 성립하지 않았다. 따라서 제조업부문의 에너지절약정책은 생산이나 수출에 역효과를 발생하지 않으면서 추진될 수 있다는 Qzturk (2010)의 보존가설을 지지하고 있다. 장기적으로는 생산, 에너지소비, 수출, 노동, 자본 간에 장기 공적분관계가 존재하며, 장기균형관계에서 에너지소비가 생산의 증가에 기여하는 것으로 나타났다. This study analyzed the mutual causal relationship between energy consumption, production, and export for manufacturing industry in Korea. The Korean manufacturing industry was divided into nine industries and panel data was constructed from 1991 to 2013. The panel Granger causality test method developed by Demitrescu and Hurlin (2012) was used along with the Vector Error Correction Model. This analysis showed that there was Granger Causality from production to energy consumption, from exports to energy consumption. However, Granger Causality was not established in the opposite direction. Therefore, this result supports the conservation hypothesis of Qzturk (2010) that energy-saving policies in the manufacturing sector can be implemented without adverse effects on production or exports in short-run. There is a long-run cointegrating relationship between production, energy consumption, exports, labor, and capital in the Korean manufacturing sector. Furthermore, the energy consumption contributes to the increasing of production in long-run equilibrium relationship.

A Research on the Economic Growth, Export, Population, and Disparity

Yeon Joon Kim 한국유럽학회 2013 유럽연구 Vol.31 No.3

기술의 발달은 산업화와 공업화를 촉진시킨다. 이는 지속 가능한 적정 인구유지에 기여하고 인구 증가는 경제 발전에 기여한다. 경제 발전이 진행되면서 또 한편으로는 국가 내지역 격차와 국가 간 지역 격차의 문제가 전 세계적으로 대두되고 있다. 한편 세계화와 국제화가 진행되면서 각국은 내수 포화 문제를 해결하기 위해 수출주도형 성장전략에 의존하게 된다. 이와 같은 국내외 경제 현황 속에서 본 연구에서는 유럽의 거시경제 데이터를 가지고 지역격차를 분석하고 수출주도형 경제성장을 테스트 하며 성장과정에서 발생되는 유럽 내 국가 간 그리고 국가 내 지역격차의 문제를 연구한다. 이와 같은 다이내믹한 거시경제 연관관계를 분석하기 위해 본 연구에서는 서베이 조사, 패널데이터 분석, 그랜저 인과분석의 세가지 테스트를 시행하여 위에서 언급한 내용들을 분석하고자 한다. 주요 분석 결과는 다음과 같다. 첫째, 본 연구에서 European Quality Life Survey 결과, 소득 측면에서 지방 임금이 도시 임금 보다 낮았고 고품질 노동력은 도시에 집중되어 있었다. 또한 교육수준은 경제성장에 그다지 영향을 미치지 않는 것으로 나타났고 저소득국가에게는 농업분야 소득이 지방에서 높았다. 둘째, 본 연구에서는 유럽 25개 국가를 대상으로 패널데이터 분석을 시행한다. 분석결과 이들 국가의 경우 인구는 경제성장에 영향을 미치는 것으로 나타났다. 따라서 본 연구는 한 국가를 효율적으로 유지하기 위해서는 지속 가능한 일정규모 이상의 인구가 유지되어야 안정적인 경제성장에 도움이 되는 것을 발견하였으며, 이는 지역 격차를 줄일 수 있는 하나의 중요한 요인으로 작용한다는 사실을 발견하였다.셋째, 본 연구에서는 유럽의 네 개의 국가를 대상으로 두 가지 그랜저 인과분석을 한다. 먼저 각 국가별 변수들 간의 그랜저 인과분석을 한다. 그 다음으로 각 국가 간 변수들 간의 그랜저 인과분석을 한다. 첫 번째 분석 결과, 수출이 경제 성장에 그리고 경제성장이 수출에 각각 그랜저 인과 관계가 있는 것으로 나타났다. 이는 적절한 수출 전략 달성으로 인해 각 국은 경제성장을 달성한다는 것을 볼 수 있다. 두 번째 분석 결과, 국가 간에도 수출과 경제 성장 간에는 밀접한 상호 연관된 그랜저 인과관계가 있는 것으로 나타났으며, 이와 같은 그랜저 인과관계는 국가 간 지역격차를 줄일 수 있는 중요한 매개체적인 역할을 하게 된다. 본 연구에서 분석한 세 가지 테스트 결과를 종합해 보면, 산업화와 공업화로 도시와 농촌 간에 지역격차가 발생하였고 또한 국가 간 지역 격차가 발생하였다. 그리고, 지속 가능한 일정 수준 이상의 인구증가가 있어야 경제성장에 기여할 수 있는 것으로 나타났다. 특히, 산업화와 공업화는 포화된 내수 문제와 더불어 수출 증대에 기여하게 되고, 수출 증대는 국가별 경제성장에 기여하게 된다. 결국 지속 가능한 인구증가와 더불어 수출 주도형경제성장으로 인해 국가 간 지역 격차는 줄어들고 있으며, 국가 내 지역 격차 문제도 같은 맥락으로 경제성장으로 줄어들고 있는 것으로 분석되었다. With the technological development and the industrialization, overall population has been grown. ‘Inter-regional disparity’ or ‘inter-regional inequality’ and ‘disparity in one country’ or ‘inequality in one country’ have been spread over the world. With the globalization and the internationalization, each country pays more attention to the export-led growth. This research investigates the regional disparity among the European countries. Also, this paper investigates one country`s disparity in the EU. This paper conducts three empirical tests. First, this paper conducts survey analysis by using the European quality life survey (EQLS). Second, this paper investigates the panel data analysis. With the panel data analysis, this paper investigates whether population affects economic growth in the European countries. And, this paper studies whether changes in population affect changes in growth rate in the European countries. Third, this paper conducts the Grangercausality test. With the analysis, this paper analyzes whether there exists the Granger-causality among economic growth, exports, and unemployment. First, with the result of the European quality life survey, the rural income is less than the urban income. And, high quality labor force is concentrated on the urban area. We find that educational level does not affect economic growth. For the low income countries, income of the agricultural sector is high. Second, this research conducts the panel data analysis using twenty-five European countries. For the panel data analysis, we use the EU`s data of GDP and population. Results of the analysis show that population affects economic growth. Therefore, this research finds that economic growth is accomplished by maintaining appropriate amounts of population. And, this is one factor that reduces the regional disparity or the regional inequality. Third, this paper conducts the Granger-causality analysis using four European countries. This paper conducts the Granger-causality analysis using macroeconomic data by country. For the Granger-causality analysis, this paper uses real GDP growth, merchandise export growth, and unemployment rate for the selected European countries. Additionally, this paper conducts the Granger-causality analysis using macroeconomic data among countries. First results of the Granger-causality test show that exports affect economic growth and economic growth affects exports. This shows that each country accomplishes economic growth by exports. Second results of the Granger-causality test show that exports and economic growth are closely related and this has a role to reduce the regional disparity. With this research, we find that we need appropriate population to maintain one country. And, the export-led growth is a growth engine in one country with the globalization and the internationalization. Therefore, the export-led growth with proper population contributes to reduce the regional disparity in Europe. We find that globalization and internationalization are drive-forces in accomplishing an equal development.

중국 KDR 가격과 원주 가격의 관계

유한수 ( Han Soo Yoo ) 아시아.유럽미래학회 2009 유라시아연구 Vol.6 No.3

This paper investigates the relationship between Chinese KDR and its underlying share issued by Huafeng Group. The underlying share is traded on HKEx(Hong Kong Exchanges and Clearing Limited) and the KDR of Huafeng is traded on KRX(Korea Exchange). Due to globalization in capital markets, cross-border capital flow is stimulated and international cross listing of securities is increased. International investors interested in foreign securities use DR widely. Because DR is based on the underlying share, the underlying share and DR may be priced in the same way. Prior studies found that the price of shares listed in foreign markets is affected by the price of their underlying shares in their home country. This study is distinguished from prior studies in that observed stock price is decomposed into fundamental value and transitory price. Some portion of stock prices is caused by noise trading of irrational traders. Therefore, observed stock price may be defined as the sum of fundamental value and transitory price. Because fundamental value is trend component, it is modelled as a random walk process. And transitory price is assumed as AR(1) process because it passes away with time. The methodological procedures used in this paper are the following steps. As a first step, to decompose observed price, state space model and Kalman filtering method are used. State space model is useful in analyzing a time series model that involves unobservable variables. And Kalman filter is the basic tool to deal with state space model. In this paper, fundamental value and transitory price are unobservable variables. Second, to check for stationarity characteristics of the series, ADF test is used. The appropriate lag order is selected with the use of the SIC. ADF test results suggest that the null hypothesis of a unit root is accepted when observed price and fundamental value are in level, and the null hypothesis of a unit root is rejected at 1% significance level when they are in first differences. If the process of series is found to be a unit root process, the next step is to check the cointegration between KDR and underlying share. If the series are found to be cointegrated, Granger causality in a VECM should be applied. If a time-series of system includes integrated variables of order one and cointegrating relations, then this system can be more appropriately specified as a VECM rather than a VAR. If there is no cointegrating relation between variables, they are to be transformed to stationary processes by differencing the series. Third, to analyze the relationship between KDR price and the underlying share price, this study employs Granger causality test. It is a useful tool to help uncover the direction of causality. Each pair of time series data has different characteristics. In the case of observed price data, two series are nonstationary and they are cointegrated, Granger causality test based on a VECM is applied. Also, in the case of fundamental value data, two series are nonstationary and they are cointegrated, Granger causality test in a VECM is applied. However, in the case of transitory prices, they do not contain trends. Because each transitory price series is stationary, Granger causality test based on a VAR is applied. This paper uses daily data running from 26 November 2007 to 27 February 2009. Because Korean stock market holidays differ from Hong Kong stock market holidays, the corresponding date which is a holiday in another country is deleted in the data set. The main empirical results of this paper are as follows. First, there exists bidirectional causalities between the observed price of the underlying share and the observed price of KDR at 1% level of significance. This means that there exists the significant linkage between two securities. This finding is similar to the results from Park and Kim(2001), Yae and Chung`s(2007) empirical findings. The impulse response function exhibits that the observed price impulse of the underlying share increases the observed price of KDR. And the observed price impulse of KDR increases the observed price of the underlying share. Second, I found that there is two way Granger causalities between the fundamental value of the underlying share and the fundamental value of KDR, similarly to the case of observed price. The trend portion of stock price is considered as the fundamental value of stock. This means that there exists the significant linkage between the trend component of the underlying share price and the trend component of KDR price. This feature, namely analyzing the trend of two securities is important for investment strategies. Similar to the case of observed price, the impulse response function of fundamental value shows that the fundamental value innovation of the underlying share increases the fundamental value of KDR. And, also, the fundamental value innovation of KDR increases the fundamental value of the underlying share. Third, on the other hand, no Granger causality exists between the pair of the transitory price of the underlying share and the transitory price of KDR. Transitory price is the portion caused by noise trading. Therefore, the noise trading of the underlying share(or KDR) cannot influence the noise trading of KDR(or underlying share). A useful extension of this study is to analyze the relationship between the underlying share and DR by another refined statistical technique estimating fundamental value and transitory price.

가상화폐는 주가지수 및 금 가격에 유효한 영향을 주는가? COVID-19 전후를 중심으로

이상원,정찬식 한국경영컨설팅학회 2021 경영컨설팅연구 Vol.21 No.4

This study examines whether cryptocurrency (virtual currency) is valid as an alternative asset for stocks through analysis of its influence on virtual currency returns, stock index returns, and gold prices using daily data from January 2015 to April 2021. In addition, the changes in the influence of these assets’returns are compared by dividing them before and after the outbreak of COVID-19. The main results of this research are as follows. First, virtual currency Bitcoin and Ethereum were complementary assets. In other words, as the former rate of return increased, the latter rate of return also showed an increasing trend. Second, as a result of exploring the Granger causality, there was an interactive Granger causality between the two virtual currencies (Bitcoin and Ethereum). And this was established both before and after the Covid-19 Pandemic. Next, before the COVID-19 Pandemic, Bitcoin returns Granger-causes the KOSPI index returns and not the other way around, but after the COVID-19 Pandemic, the KOSPI returns Granger-caused the Bitcoin returns, and not the other way around. On the other hand, in the case of Ethereum and KOSPI returns, there was both-way Ganger causality, whether before or after the Covid-19 Pandemic. Meanwhile, before the COVID-19 Pandemic, the two cryptocurrency returns Granger-caused gold price returns, but not the other way around. However, after the COVID-19 Pandemic, the gold price returns Granger-caused KOSPI returns, and the cryptocurrency returns no longer affected the gold price returns. The reversal of Granger-Causality before and after the Covid-19 Pandemic is believed to be due to the serious structural change in the asset market. Third, as a result of the impulse response function analyses, the shock to the KOSPI return and gold price change rate, and the shock to the cryptocurrency itself, caused relatively large fluctuations in the KOSPI returns and gold price returns within 20 days. However, this phenomena were mostly driven by the influence before the Covid-19 Pandemic, and after the Covid-19 Pandemic, the impact on each variable had a longer (50 days) and greater impact on other variables. In short, the two virtual currencies, Bitcoin and Ethereum, are complementary to each other, and the structural change in their returns after the Covid-19 Pandemic has shown a longer response to small shocks in economic variables, resulting in increased sensitivity. 본 연구는 가상화폐가 주식의 대체자산으로 유효한지 2015년 1월부터 2021년 4월까지의 일별자료를 이용하여 가상화폐와 주식 및 금 상호간에 미치는 영향력 분석을 통해 알아본다. 또한 코로나 19 발생 이전과 이후로 구분하여 이들 상품 상호간의 영향력의 변화에 대해 비교한다. 연구 분석의 주요 결과는 다음과 같다. 첫째, 가상화폐 비트코인과 이더리움은 보완재 성격의 자산이었다. 즉, 전자의 수익률이 증가할 때 후자의 수익률 역시 증가하는 추세를 보였다. 둘째, 그랜저 인과관계를 탐색한 결과, 두 가상화폐(비트코인, 이더리움) 간에는 쌍방향의 그랜저 인과관계가 존재하였다. 그리고 이는 Covid-19 Pandemic 이전과 그 이후 공히 성립하였다. 다음으로, Covid-19 Pandemic 이전에는 비트코인 수익률이 코스피지수 수익률을 Granger-Cause 하는 대신 그 역은 성립하지 않았으나, Covid-19 Pandemic 이후에는 상황은 전반대가 되어 코스피지수 수익률이 비트코인 수익률을 Granger-Cause 하는 대신 그 역은 성립하지 않았다. 반면 이더리움의 경우 Covid-19 Pandemic 이전이건 이후이건 일관되게 쌍방향의 인과관계가 나타났다. 한편, Covid-19 Pandemic 이전에는 두 가상화폐 수익률이 금가격 변화율을 Grand-Cause 하였으나 그 역은 성립하지 않은 대신, Covid-19 Pandemic 이후에는 상황이 역전되어 금가격 변화율이 가상화폐 수익률을 Grand-Cause 할 뿐, 가상화폐 수익률은 더 이상 금가격 변화율에 영향을 미치지 않았다. 이처럼 Covid-19 Pandemic 이전과 이후 기간에서의 Granger-Causality의 역전 현상은 Covid-19 Pandemic이 자산 시장에 심각한 구조적 변화(Structural Change)를 가져왔기 때문인 것으로 판단된다. 셋째, 충격반응함수 분석 결과, 코스피지수 수익률 및 금가격 변화율에 대한 충격은, 그리고 가상화폐 자기 자신에 대한 충격은, 두 가상화폐 수익률에 20일~30일까지 영향을 미치면서 상대적으로 큰 수익률의 변동을 가져왔으나, 코스피지수 수익률과 금가격 변화율에는 20일 이내로만 영향을 미쳤을 뿐만 아니라 상대적으로 작은 수익률 변화를 가져왔다. 그런데 이러한 현상은 대부분 Covid-19 Pandemic 이전의 영향력에 의해 주도된 것이며, Covid-19 Pandemic 이후에는 각 변수에 대한 충격은 여타 변수에 대하여 보다 더 오랫동안 (50일 이상), 그리고 더욱 더 큰 영향을 미쳤다. 요컨대, 비트코인 및 이더리움 두 가상화폐는 상호 보완재이며, Covid-19 Pandemic 이후 그 수익률에 있어 구조적 변화(structural change)를 겪음으로써 경제변수의 작은 충격에도 더욱 큰 반응을 더 오랫동안 보이는, 민감도 상승의 효과를 보였다.

산업생산지수 변동성과 외국인 직접투자의 선도-지연 관계

유한수 ( Han Soo Yoo ) 아시아.유럽미래학회 2011 유라시아연구 Vol.8 No.1

Foreign direct investment(FDI) not only increases output and employment, but also transfers advanced technologies. That is, FDI may promote economic growth and enhance competitiveness. One of the most important determinants of FDI is the market size. Larger market size offers opportunities to realize economies of scale. Market size is a proxy for product demand the potential for growth. This study uses industrial production index(IPI) as a proxy to market size, and examines the relationship between industrial production index volatility and foreign direct investment. That is, this paper investigates whether the uncertainty of real economy activity of the host country influences the decision-making of foreign investors. Previous studies on this subject focus mainly on the relationship between IPI in level and FDI. On the contrary, this study is distinguished from prior studies in that it investigates the relationship between IPI volatility and FDI. When decisions have an irreversible factor, uncertainty about future outcomes plays a key role in the decision to invest. Because FDI generally involves irreversible costs in the foreign country, foreign companies wait for a more favorable economic environment before investing abroad. Therefore, irreversibility of investment expenditures renders the investment decisions of foreign companies sensitive to economic uncertainty. The descriptive statistics show excess kurtosis, and the Jarque-Bera test statistic rejects the null hypothesis of normality for the unconditional distribution of the monthly IPI changes. To test ARCH effects in the residuals, we employ ex ante ARCH LM test. The null hypothesis of ARCH LM test is that there is no ARCH up to q order in the residuals. This is a regression of the squared residuals on a constant and lagged squared residuals up to order q. Ex ante ARCH LM tests show that there are ARCH effects in the residuals. Therefore, GARCH type model is appropriate for analyzing this data. The procedure used in this paper involves the following steps. First, to determine IPI volatility, I use the exponential generalized autoregressive conditional heteroskedasticity model proposed by Nelson. EGARCH model has a distinctive feature, namely, conditional variance is modeled to capture the asymmetry effect of volatility. The return series of IPI shows the serial correlation. This causes the distortion of parameter estimates in GARCH type model. To account for serial correlation in the returns, I include a MA(1) process for the residuals in the mean equation. According to the results, the asymmetric volatility effect is significant at 5% level. In this study, the asymmetry effect term is negative and statistically different from zero, indicating that the IPI volatility impact is asymmetric during the sample period. The economic implication of this sign of the leverage effect term is that a decrease in IPI would lead to a higher level of uncertainty when compared to the level of uncertainty generated by an increase in IPI. In estimated EGARCH model, the persistence parameter which is the coefficient of lagged conditional variance in the variance equation is smaller than one. Therefore, estimated conditional variance series is stationary. Expost LM statistics test whether the residuals exhibit ARCH effect. If the EGARCH model is correctly specified, there should be no ARCH effect in the residuals. P-value indicates that there is not any ARCH effect up to order 5. Therefore, the EGARCH model is well specified. Second, a prerequisite in applying the Granger causality test is to test the unit root properties of the series. To examine the stationarity of the series, the Augmented Dickey-Fuller test is applied. The optimum lag order is determined by the SC. ADF test results show that the null hypothesis of a unit root is rejected in cases of IPI volatility and FDI, respectively. That is, IPI volatility and FDI are stationary series. Therefore, raw data is used in this study. Third, to examine the lead-lag relationship between IPI volatility and FDI over the data from January of 1999 to October of 2010, this study employs the Granger causality test. Granger causality test is a useful statistical technique to examine the direction of causality. One of the steps in Granger causality test is determining the lag length included in the model for each pair of series. This step is important due to a high sensitivity of VAR to model specification. The optimum lag length is determined by the SC. Since IPI volatility and FDI are found to be stationary series, Granger causality test based on VAR is applied. The Granger causality tests show that there is one-way Granger causality between IPI volatility and FDI. The null hypothesis that IPI volatility Granger causes FDI is rejected, implying that causation is running from IPI volatility to FDI. Fourth, to analyze the time path of the IPI volatility shock on FDI, the impulse response function analysis is employed. The impulse response function shows the response that a shock to the innovation of one variable has on the other variables in the model. The IRF shows that FDI initially decreases in response to a shock to IPI volatility and reaching its trough in the second month. The negative response of FDI continues to the third month. That is, IPI volatility shock decreases the magnitude of FDI over the three months. In conclusion, these findings suggest that stabilizing IPI volatility will help increase FDI. The implication to policy makers from this study is that stabilizing IPI volatility is important for encouragement of FDI inflow. Direction for further research is to investigate the relationship between IPI volatility and FDI by industry.

A Research on the Economic Growth, Export, Population, and Disparity:in case of the EU

김연준 한국유럽학회 2013 유럽연구 Vol.31 No.3

With the technological development and the industrialization, overall population has been grown. ‘Inter-regional disparity’ or ‘inter-regional inequality’ and ‘disparity in one country’ or ‘inequality in one country’ have been spread over the world. With the globalization and the internationalization, each country pays more attention to the export-led growth. This research investigates the regional disparity among the European countries. Also, this paper investigates one country’s disparity in the EU. This paper conducts three empirical tests. First, this paper conducts survey analysis by using the European quality life survey (EQLS). Second, this paper investigates the panel data analysis. With the panel data analysis, this paper investigates whether population affects economic growth in the European countries. And, this paper studies whether changes in population affect changes in growth rate in the European countries. Third, this paper conducts the Grangercausality test. With the analysis, this paper analyzes whether there exists the Granger-causality among economic growth, exports, and unemployment. First, with the result of the European quality life survey, the rural income is less than the urban income. And, high quality labor force is concentrated on the urban area. We find that educational level does not affect economic growth. For the low income countries, income of the agricultural sector is high. Second, this research conducts the panel data analysis using twenty-fiveEuropean countries. For the panel data analysis, we use the EU’s data of GDP and population. Results of the analysis show that population affects economic growth. Therefore, this research finds that economic growth is accomplished by maintaining appropriate amounts of population. And, this is one factor that reduces the regional disparity or the regional inequality. Third, this paper conducts the Granger-causality analysis using four European countries. This paper conducts the Granger-causality analysis using macroeconomic data by country. For the Granger-causality analysis, this paper uses real GDP growth, merchandise export growth, and unemployment rate for the selected European countries. Additionally, this paper conducts the Granger-causality analysis using macroeconomic data among countries. First results of the Granger-causality test show that exports affect economic growth and economic growth affects exports. This shows that each country accomplishes economic growth by exports. Second results of the Granger-causality test show that exports and economic growth are closely related and this has a role to reduce the regional disparity. With this research, we find that we need appropriate population to maintain one country. And, the export-led growth is a growth engine in one country with the globalization and the internationalization. Therefore, the export-led growth with proper population contributes to reduce the regional disparity in Europe. We find that globalization and internationalization are drive-forces in accomplishing an equal development.

일본 엔 가치의 변화가 KOSPI와 NIKKEI 225의 상호영향력 변화에 관한 실증적 연구

임병진,김종택 한국일본근대학회 2015 일본근대학연구 Vol.0 No.49

이 연구는 엔의 가치변화가 KOSPI와 NIKKEI 225에 미친 상호영향력에 관한 실증적 연구로 2004년 1월 3일 부터 2015년 2월 25일 까지 583개의 주간자료를 사용하여 VAR모형과 VAR모형을 이용한 예측오차의 분산분해기법과 충격함수분석 및 Granger인과관계(Granger causality)검정을 통하여 인과관계와 상호영향력을 분석하였다. 본 연구의 중요한 결과들을 요약하면 다음과 같다. 첫째, 상관계수는 엔 환율과 KOSPI의 상관계수가 0.235672이고 KOSPI과 NIKKEI 225의 상관계수는 0.096753으로 나타났고, 엔 환율과 NIKKEI 225의 상관관계는 -0.892009로 나타났다. 둘째, 엔 환율과 KOSPI 및 NIKKEI 225 자료의 수준변수는 불안정적으로 나타났으나, 로그 차분 후에는 안정적인 시계열로 나타났다. 셋째, 엔 환율과 KOSPI 및 NIKKEI 225 시계열 자료간의 공적분 관계가 있는 것으로 나타났다. 넷째, 엔 환율과 KOSPI간에는 1%의 유의수준에서 Granger 인과관계가 있는 것으로 나타났다. 또한 KOSPI와 NIKKEI 225간에도 역시 1%의 유의수준에서 Granger 인과관계가 있는 것으로 나타났고 NIKKEI 225과 엔 환율액 간에도 1%의 유의수준에서 Granger 인과관계가 있는 것으로 나타났다. This study is an empirical study on the changes in the value of the yen mutual influence of KOSPI and NIKKEI 225. In this paper, 583 weekly data from January 3, 2004 to February 25, 2015 were used. ADF(Augmented Dickey-Fullerand and PP(Phillips and Perron) tests, co-integration test, Granger causality test, impulse response analysis, variance decomposition analysis and autoregressive(VAR) model were employed. In summary, the important findings of this study are as follows. First, the correlation coefficient yen exchange rate and KOSPI and KOSPI’s correlation coefficient of .235672 and NIKKEI 225 Watering the matter appeared to .096753, the correlation of the yen exchange rate and the NIKKEI 225 was found to be -0.892009. Second, the yen exchange rate and KOSPI and the level of the NIKKEI 225 data variables or’ve shown as unstable, after the log difference was found to be a reliable time series. Third, the yen showed that the co-integration relationship between the exchange rate and KOSPI and NIKKEI 225 time-series data. Fourth, the yen exchange rate and KOSPI between 1% significance level showed that the Granger causality. In addition, even between KOSPI and NIKKEI 225 also showed that there is Granger causality at the significance level of 1% NIKKEI 225 yen exchange fluid and showed that even between the Granger causality at the significance level of 1%.

재정과 경제성장 간 인과관계: 중국의 사례

유듕국 ( Dong Gug Yoo ),김일태 ( Il Tea Kim ),송명화 ( Ming Hua Song ) 한국경제통상학회 2011 경제연구 Vol.29 No.3

본 논문은 중국의 1978년-2008년 자료로 재정과 경제성장의 Granger 인과관계분석을 통해서 일인당 국내총생산과 조세수입, 재정지출의 장·단기 인과관계를 분석하고 있다. 분석결과를 요약하면 먼저, 중국 국내총생산과 조세수입 간에 양방향으로 Granger 인과관계가 존재한다. 둘째, 일방적인 인과관계는 재정지출에서 국내총생산 방향, 재정지출에서 조세수입 방향으로 존재한다. 셋째, 충격반응분석에서는 국내총생산의 충격과 조세수입의 충격이 재정지출에 부(-)의 효과를 보이고 있다. 마지막으로, 장기인과관계분석에서는 국내총생산과 조세수입 간에 현실적인 장기균형이 존재하지 않음을 보여준다. This paper estimates Granger causality test between public finance and economic growth using from 1978 to 2008 data in China in the context of vector error correction model. This paper also examines both short and long-term causality between gross domestic product and public finance such as fiscal expenditures and tax revenues. This paper shows that tax revenues and fiscal expenditure have Granger-causes with gross domestic product in China and that gross domestic product have Granger-causes with tax revenues and fiscal expenditure have Granger-causes with tax revenues. This paper also reveals that impulse responses of gross domestic product and tax revenue in China decrease gross domestic product, respectively. There does not exists long-run equilibrium relationship between the tax revenue and gross domestic product in China.