Font Size: a A A

Inter-Industry Long-and Short-Position Risk Contagion Effect In The Chinese Stock Market

Posted on:2022-11-12Degree:DoctorType:Dissertation
Country:ChinaCandidate:Y T LiuFull Text:PDF
GTID:1489306749463324Subject:Finance
Abstract/Summary:PDF Full Text Request
Industries are important economic link to connect the macro economy and the micro economy.With the rapid development of China's economy,the inter-industry business penetration and mixed business operation is gradually becoming a trend,so the industry-specific risks are transmitted easier among related industries.Research on industry risk measurement and the inter-industries risk contagion effect is gaining extensive attention.The stock market accelerates capital flow,which supports the industry development and makes resources allocated more efficiently across industries.Meanwhile,the stock market's influence on different industries has also deepened.Stock market is a barometer of macroeconomic,to a large extend,the industrial stock indices reflect the actual development of each industry.Therefore,the most direct way to understand the risk contagion effect across industries is to analyze the risk spillovers between industrial stock indices.In the stock market,investors in long position expect the stock prices rise,so they face the risk of stock prices falling.In contrast,investors in short position expect the stock prices fall,they face the risk of stock prices rising.Therefore,the risks faced by investors in long and short positions are asymmetrical,this fact that is reflected in the asymmetrical characteristics of distribution of industrial stock indices in the Chinese stock market,such as “leptokurtic”,“fat tailed” and “left-skewed”.The left tail of the stock index return distribution corresponds to the risk of a fall in the stock,i.e.long risk,while the right tail corresponds to the risk of a rise in the stock price,i.e.short risk.Hence,studies related to the contagion effect of risk between industries need to be conducted on the premise that long risk is asymmetric to short risk.However,although many scholars have studied the inter-industry risk contagion effect,there's few literatures have been found distinguishing between long-and short-positions to study the inter-industry risk contagion effect in the stock market.Starting from the long-and short-position risks,this paper divide short-,mediumand long-term investment horizons,different market states and five severely volatile periods of the stock market to investigate the dynamic changes of inter-industry risk contagion effects in both time and frequency domains.The specific steps can be summarized as follows.The first step is to measure the industry risk in short and long positions.In considering the possible impact of stock market on the industry return series,the study first extracts industry-specific return series by using the Capital Asset Pricing Model(CAPM).Based on the industry-specific returns,then measures the industry Va R series of industries in short and long positions by 7 parametric and non-parametric GARCH models,and tests the accuracy of the measured industry Va R series using three posthoc tests of industry risk measurement: unconditional coverage test,conditional coverage test and dynamic quantile regression test.The results show that the nonparametric CAVia R indirect GARCH model exhibits more accurate measurement of industries' risks under all three test criteria compared to other parametric GARCH models.In addition,the industry-specific risk series obtained by CAVia R-IG model exhibit asymmetric characteristics.The next step is distinguishing short-,medium-and long-investment horizons to study the dynamics of inter-industry risk contagion effects.Since investors have short-,medium-and long-investment horizons and the inter-industry risk contagion effects for different investment horizons is time-varying,it is necessary to discuss the dynamics of the inter-industry risk contagion effect at different investment horizons.In this paper,we employ the Bayesian partial likelihood estimation method(QBLL)to estimate the dynamic neighborhood variance matrix of the TVP-VAR model and then decompose this variance matrix by using spectral decomposition method.Unlike previous studies,this method can analyze the risk spillovers in the full sample without resorting to the rolling window technique,and it is able to demonstrate the interindustry risk contagion effect over time at different investment horizons.After careful comparison,it is found that the overall inter-industry risk contagion effect in China's stock market rises during severely volatile periods of the stock market,and the short-,medium-and long-inter-industry risk contagion exist significant differences.The results show that,firstly,there is a significant risk contagion effect between industry stock indices,and the short-term inter-industry risk contagion effect is strongest,followed by the medium term and weakest in the long term.Secondly,during stock market severely volatile periods,the inter-industry risk contagion in the short term decreases rather than rises,while inter-industry risk contagion in the medium and long term appears to increase significantly.Third,we examine the correlation between the strength of inter-industry risk contagion effects and volatility cycle in the Chinese stock market,and to extend,diagnose which industry have asymmetric risk contagion characteristics.Combined with different stock market volatility phases(which include bull,bear and shock market phases)and the average values of total inter-industry risk spillovers,the results found that the strength of inter-industry risk contagion effect have little correlation with changes in market phases.In addition,after analyzing the characteristics of interindustry net risk spillovers at different market phases,the results inferred that the risk spillovers of energy,industrial and consumer staples exhibit significant asymmetric characteristics both in amplitude and directions,while in other industries,the asymmetric characteristics only evidenced by the changing amplitude of risk spillovers.Finally,given that the structure of the inter-industry risk contagion effect changes significantly during stock market severely volatile periods,this paper then draws 40inter-industry risk contagion networks of five representative market volatile periods in both time domains and frequency domains,and summarizes the regularity and variability of these networks.The results indicate that the industrial sector is more risklinked with other sectors and tend to develop into risk source during stock market severely volatile periods,while the consumer staples sector is less risk-linked to other sectors and are defensive industries.From the perspective of different investment horizons and market states,this paper examines the dynamic characteristics of long-and short-position inter-industry risk contagion effect in China's stock market both in time and frequency domains.The empirical results of this paper extend the research of inter-industry risk contagion effects and obtain many novel conclusions.which provide a new perspective for analyzing inter-industry risk contagion.The main findings of this paper suggest that,firstly,there is a highly significant risk contagion effect among industries in China's stock market,which indicate that specific industry risks are highly susceptible to spread to its associated industries.Secondly,the risk contagion effect is asymmetric in the energy,industrial and consumer staples sectors,this suggest that stock market regulators and policy makers should differentiate between "positive" and "negative" news when they are managing stock market risks and making relevant policies.Stock market investors also need to differentiate long-and short-positions when allocating and diversifying their risk exposure to stocks in these industries.Thirdly,during severely volatile periods of the stock market,the short-term risk contagion effect between industries is decreasing,while the medium-and long-term risk contagion effect is significantly enhanced.Therefore,during severely volatile periods of the stock market,a portfolio with a shortterm investment horizon can obtain better risk avoiding effect than that of a portfolio with a medium-and long-term investment horizon.The contents of this paper are in line with the policy guidelines of the Party Central Committee on economic and financial work in the new era,and the results of this paper are an important theoretical reference and practical reference for the prevention and resolution of systemic financial risks,as well as the healthy and orderly development of the stock market.
Keywords/Search Tags:Inter-industry risk contagion, Spectral decomposition of TVP-VAR model, Long-and short position VaR, Different investment horizons, Different market states
PDF Full Text Request
Related items