주식수익률 및 변동성의 업종간 전이효과 분석

Abstract

The study on the spillover effect of industrial return and volatility

Hee Jin Lee

Department of Economics, The Graduate School, Pukyong National University

Abstract

Participants into the financial market have interest in the changes in each index by industry and relations between different types of industry because of various reasons. Fund managers, institutional investors and personal investors use indexes by industry as a benchmark. In particular, portfolio managers use them to judge which industry and how much they will choose for their portfolio. Accordingly, it is important not only to predict the profits by industry and but also to understand the relations between different types of industry. This work used generalized forecast error variance decomposition model to analyze the short-term movement of spillover phenomena of and stock return and volatility in KOSPI. As a result, important findings were identified as follows. First, it was found that return of each industry index and the short-term flow of volatility were similar and synchronized. Spillover effect of each return of 14 types of industry during the entire period (Jan. 4, 1990 to Aug. 29, 2012) was analyzed and was found to be 83.8%. The result was similar in the case of use of weekly data (84.9%). It was also similar to the result drawn from the analysis of the post-2000 year data including four additional types of industry (daily data: 84.2%; weekly data: 84.1%). The spillover effect of each variance of 14 types of industry during the entire period (Jan. 4, 1990 to Aug. 29, 2012) was analyzed and was found to be 75.0%. In the case of the use of weekly data, it was 63.1%. Therefore, it was found that there was considerable spillover effect in volatility by industry. In the analysis of post-2000 year data including four additional types of industry (daily data: 82.0%; weekly data: 71.9%), the spillover effect of volatility was found. Secondly, each spillover effect of return and volatility was found to have time variability. Each spillover effect of return and volatility by industry receives bad influence during the time when the stock market faces bad conditions and falls down so that the spillover effect is expanded to each type of industry. But, at the time when the stock market faces good conditions and is brisk, differentiation phenomenon by industry appears, and thus the spillover effect was found to drop. Lastly, the leading types and following types changed at the lapse of time. The net spillover effect that is the difference between the influence given to other types of industry and the influence received from other types of industry was examined through recursive regression. As a result, it was found that the leading types and following types of industry changed according to the market conditions. The results of this work are considered to be useful information for portfolio managers who should make a decision on investment in consideration of the investment weight by industry, and for stock investors who need to consider the investment of sector fund.