This study presents a comprehensive and innovative analysis of dynamic tail risk in the Chinese stock market utilizing the localizing conditional autoregressive expectiles(LCARE)model.We consider the dynamic changes i...This study presents a comprehensive and innovative analysis of dynamic tail risk in the Chinese stock market utilizing the localizing conditional autoregressive expectiles(LCARE)model.We consider the dynamic changes in the tail distribution of stock market returns and the associated time-varying parameters,which is relatively rare in existing literature on tail risk in the Chinese stock market.We first determine homogeneous intervals through the local parametric approach(LPA)and then establish a CARE model with constant parameters within the homogeneous intervals.The lengths of the homogeneity intervals obtained through LPA provide strong evidence for the presence of potential structural changes in tail risk measurement.The efficacy of the LCARE model in predicting outcomes at various time scales has also been demonstrated effectively.The empirical evidence on portfolio strategies shows that the time-invariant portfolio protection(TIPP)strategy with time-varying multipliers,which is grounded in the LCARE framework,exhibits enhanced performance in comparison to other strategies.Thus,this study has the potential to serve as a valuable reference for government departments and investors seeking to assess and alert to the time-varying tail risk of the stock market across various market conditions and investment horizons.展开更多
基金Supported by the Natural Science Foundation of Shandong Province(ZR2023MG037)the National Natural Science Foundation of China(72171192)。
文摘This study presents a comprehensive and innovative analysis of dynamic tail risk in the Chinese stock market utilizing the localizing conditional autoregressive expectiles(LCARE)model.We consider the dynamic changes in the tail distribution of stock market returns and the associated time-varying parameters,which is relatively rare in existing literature on tail risk in the Chinese stock market.We first determine homogeneous intervals through the local parametric approach(LPA)and then establish a CARE model with constant parameters within the homogeneous intervals.The lengths of the homogeneity intervals obtained through LPA provide strong evidence for the presence of potential structural changes in tail risk measurement.The efficacy of the LCARE model in predicting outcomes at various time scales has also been demonstrated effectively.The empirical evidence on portfolio strategies shows that the time-invariant portfolio protection(TIPP)strategy with time-varying multipliers,which is grounded in the LCARE framework,exhibits enhanced performance in comparison to other strategies.Thus,this study has the potential to serve as a valuable reference for government departments and investors seeking to assess and alert to the time-varying tail risk of the stock market across various market conditions and investment horizons.
