用C-Vine Copula进行预测
我有一个具有4个不同变量的数据集：X1，X2，X3和K1。我想使用C-Vine Copula对相互依赖的btw建模这4个变量，而K1是根节点，然后进行k_ {t+1}的预测。我读了几篇文章，但找不到方法。你对我有什么建议吗？
注意：顺便说一句，我还尝试在rstudio中使用rvinStructuruselect，但无法定义根节点，这将自动选择。
I have a dataset with 4 different variables: x1, x2, x3 and k1. I'd like to use C-Vine copula to model interdependency btw these 4 variables with k1 being root node, then make forecasting of k_{t+1}. I've read several articles but could not find a way. Do you have any suggestions for me?
Note: By the way, I also tried to use RVineStructureSelect in RStudio, but cannot define the root node, this is automatically selected.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
rvinStructuresElect根据Kendall的Tau值自动选择藤蔓类。要做出C-Vine选择，您应该制作函数参数type = 1。然后，这将返回您的C-vine结构，但也许还有另一个根节点以外的k1。要使用root节点k1制作C-vine结构，然后您应该手动执行此操作，然后，您不能使用rvinSternuctuctureceRucturecelectelect函数。相反，您可以使用rvinecopulaselect或rvinesequentialest。RVineStructureSelectselects the vine class automatically, based on the Kendall's tau values. To make a C-vine choice, you should make the function argumenttype = 1. Then this will returns you the C-vine structure but maybe with another root node thanK1. To make a C-vine structure with a root nodek1, then you should do this manually, then, you cannot use theRvineStructureSelectfunction. Instead you can useRvineCopulaSelectorRVineSequentialEst.