可参考Modelling Financial Time Series with S-Plus 2ed

Chap 13 Multivariate GARCH Modeling

利用function mgarch()很容易运行的.

其中formula.var 有很多可选择,

譬如:

~dvec(p,q), the basic diagonal vec model

~ewma1, exponentially weighted model of type 1

~bekk(p,q), the BEKK model

~egarch(2,2). Equivalent to ~diag.g(p,q).

~ccc.g(p,q), the conditional constant correlation model

~prcomp.g(p,q), the principal components model

.....

谢谢楼上的朋友。

恩，现在发现需要利用的是多变量VAR-EGARCH模型，这在多变量EGARCH模型上该如何处理呢？

mgarch命令只能计算不变相关系数的情况。

在加载finmetrics module 后，在commands里输入

hp.ibm <- seriesMerge(hp.s, ibm.s)

hp.ibm.bekk = mgarch(hp.ibm~1, ~bekk(1,1))

出来的是：Problem in mgarch(hp.ibm ~ 1,  ~ bekk(1, 1)): Couldn't find a function definition for "stop.on.bdObject"
Use traceback() to see the call stack
再输入traceback()

5: eval(action, sys.parent())
4: doErrorAction("Problem in mgarch(hp.ibm ~ 1,  ~ bekk(1, 1)): Couldn't find a function definition for \"stop.on.bdObject\"", 1000)
3: mgarch(hp.ibm ~ 1,  ~ bekk(1, 1))
2: eval(expression(hp.ibm.bekk = mgarch(hp.ibm ~ 1,  ~ bekk(1, 1))))
1:
Message: Problem in mgarch(hp.ibm ~ 1,  ~ bekk(1, 1)): Couldn't find a function definition for "stop.on.bdObject"

请问这是什么原因？我刚学s-plus

[此贴子已经被作者于2008-6-21 15:24:41编辑过]