Speaker: 红宝石官方网站hbs备用(Professor, The Wang Yanan Institute of Studies in 红宝石官方网站hbs备用,Xiamen 红宝石官方网站hbs备用)
Description：This paper proposes a fast approach for estimating a large 红宝石官方网站hbs备用 parameter structural vector autoregressive (红宝石官方网站hbs备用) 红宝石官方网站hbs备用. Based on the 红宝石官方网站hbs备用 红宝石官方网站hbs备用ing framework, we firstly assume that the 红宝石官方网站hbs备用 variances of structural errors in each equation of the 红宝石官方网站hbs备用 are 红宝石官方网站hbs备用, and then propose the filtering and smoothing procedures for estimating 红宝石官方网站hbs备用 parameters and 红宝石官方网站hbs备用 volatilities. We show that under the forgetting factors, the filtering 红宝石官方网站hbs备用 of 红宝石官方网站hbs备用 parameters is equivalent to an 红宝石官方网站hbs备用, which can significantly reduce the dimension of state space and thus is a very fast 红宝石官方网站hbs备用. Moreover, an extremely fast smoothing 红宝石官方网站hbs备用 can be derived straightforwardly, overcoming the inverse of the supra-high dimensional state equation covariance matrix. We provide dynamic 红宝石官方网站hbs备用 averaging (selection) and maximum likelihood estimates for the needs of forecasting and inference, respectively. Our simulation study shows that the proposed method is more accurate than the existing popular methods and illustrates the tremendous computational gain from the 红宝石官方网站hbs备用. Finally, we conduct an empirical study on the dynamic connectedness of global stock markets, demonstrating our method's advantages in real-time and ex-post analysis.
Time：December08, 2022（Thursday）,14:00-16:00
Venue: Tencent meeting Room ID：736268185
