发帖
雷达卡
各位大神你好！有一些关于用stata做dcc garch模型的具体操作问题想请教一下，求帮忙谢谢！
to examine the hedging capability of Financial asset A against others. There are two variable steps in this procedure for this model:The first order univariate GARCH models are estimated :
-The mean equation: rt = c + Q rt + et-The variance equation: ht = c + W e^2t-1 + X ht-12.
The correlation component ρt is maximized in the second step:
ρt = ( 1−α−β) ρ + α(εt−1)(έt−1) + βρt−1
Then the question come as I want to extract the time varying correlations ρt from the above model into a separate time series for each sector.
ρt are regressed on dummy variables representing market turmoil to test Stock A as a hedge and safe haven asset against stock sector risk:
ρt = γ0 + γ1D (rstockq10) +γ2D (rstockq5) + γ3D (rstockq1)
where D represent dummy variables that capture extreme movements in the underlying stock sectors at the 10%, 5%, and 1% quantiles of the most negative stock returns.
我输入这命令对吗？
mgarch dcc (XBTUSDBGNLCurncyR1 BRLUSDCurncyL3 RUBUSDCurncyR2 INRUSDCurncyL1 CNYUSDCurncyL2 ZARUSDCurncyR3 =, arch(1) garch(1))
得出dcc的图表后如何用取出correlation的数据和应该输入什么命令做这个回归分析？
ρt = γ0 + γ1D (rstockq10) +γ2D (rstockq5) + γ3D (rstockq1)
万分感谢!!
京公网安备 11010802022788号
