A1697
Title: Causal effects on volatility by causal-GARCH model
Authors: Xing Wang - Durham University Business School (United Kingdom)
Haofeng Liao - Durham University Business School (United Kingdom) [presenting]
Abstract: A novel approach, causal-GARCH (C-GARCH) model is proposed to investigate the causal effect of a treatment on the volatility of financial returns within a potential outcome framework. A methodology of causal inference and hypothesis test of a causal effect on volatility is provided based on Bootstrap. The performance of the new model is tested and evaluated, further compared with the GJR-GARCH and E-GARCH models by data simulations. The results of simulations show that the C-GARCH model can precisely identify the causal effect of the intervention on the volatility of interest in a short horizon, while GJR-GARCH and E-GARCH cannot identify a certain treatment effect. For empirical application, the C-GARCH model is employed to estimate the causal effect of the Russian invasion of Ukraine on both European and US stock markets. It is found that this war has earlier and more significant causal impacts on the volatility of European stock market than the US market.
