본 연구는 실물경제부문의 불확실성을 나타내는 산업생산지수 변동성과 주식시장의 불확실성을 나타내는 KOSPI 변동성 간의 관계에 대해 1990년 1월부터 2008년 11월까지의 월별 자료를 이용하여 분석하였다. 기존 연구들은 수준변수들 간의 관계 또는 GARCH류의 모형을 이용하여 계산한 관측된 변동성 간의 관계만을 분석하였으나 본 연구에서는 관측된 변동성을 상태공간모형을 이용하여 영속적 변동성과 일시적 변동성으로 분해하여 보다 심도 있는 분석하였다는 데에서 기존 논문들과 차별을 두었다. 분석 결과, 산업생산지수의 관측된 변동성과 KOSPI의 관측된 변동성 간의 관계, 산업생산지수의 일시적 변동성과 KOSPI의 일시적 변동성 간의 관계에서는 서로 그랜저 인과관계가 없는 것으로 나타났으나, 영속적 변동성의 경우는 KOSPI의 영속적 변동성이 산업생산지수의 영속적 변동성에 대해 선도관계를 갖는 것으로 나타났다. 이와 같은 분석결과는 실물부문의 안정적인 추세를 위해서는 주가지수 추세의 안정이 필요하다는 경제적 함의로 해석될 수 있다.

The objective of this study is to investigate the relationship between Industrial Production Index(IPI) volatility and KOSPI volatility. That is, this paper elucidates the relation between the volatility of real economy activity and stock market volatility. Because stock price is the present value of expected cashflows of corporation, the volatility of real economic activity may be related to stock price volatility. Previous studies have been concentrated on the relationship between observed volatility of IPI and observed volatility of stock price index. But the analysis in this study is distinguished from previous studies in that observed volatility is decomposed into permanent volatility and transitory volatility. The first step in this paper is to estimate observed volatility by EGARCH model. In the second step, I use state space model to decompose observed volatility. Permanent volatility is modelled as a random walk process and transitory volatility is assumed as a stationary process. In the third step, the ADF test is used to check the stationarity of the volatility series data. In the fourth step, Granger causality test is used to examine the relationship between IPI volatility and KOSPI volatility. If a set of variables are cointegrated, Granger causality test in a VECM is employed. The findings of this paper are as follows. First, no Granger causal relations exist between the pair of IPI observed volatility and KOSPI observed volatility, and the pair of IPI transitory volatility and KOSPI transitory volatility. Second, Granger causation exists from permanent volatility of KOSPI to permanent volatility of IPI from lag 3 to lag 6.

The objective of this study is to investigate the relationship between Industrial Production Index(IPI) volatility and KOSPI volatility. That is, this paper elucidates the relation between the volatility of real economy activity and stock market volatility. Because stock price is the present value of expected cashflows of corporation, the volatility of real economic activity may be related to stock price volatility. Previous studies have been concentrated on the relationship between observed volatility of IPI and observed volatility of stock price index. But the analysis in this study is distinguished from previous studies in that observed volatility is decomposed into permanent volatility and transitory volatility. The first step in this paper is to estimate observed volatility by EGARCH model. In the second step, I use state space model to decompose observed volatility. Permanent volatility is modelled as a random walk process and transitory volatility is assumed as a stationary process. In the third step, the ADF test is used to check the stationarity of the volatility series data. In the fourth step, Granger causality test is used to examine the relationship between IPI volatility and KOSPI volatility. If a set of variables are cointegrated, Granger causality test in a VECM is employed. The findings of this paper are as follows. First, no Granger causal relations exist between the pair of IPI observed volatility and KOSPI observed volatility, and the pair of IPI transitory volatility and KOSPI transitory volatility. Second, Granger causation exists from permanent volatility of KOSPI to permanent volatility of IPI from lag 3 to lag 6.

industrial production index, stock price index, permanent volatility, transitory volatility, state space model