Multivariate time series exhibit two types of dependence: across variables and points. Vine copulas are graphical models for the dependence can conveniently capture both in same model. We derive maximal class graph structures that guarantee stationarity under a natural verifiable condition called translation invariance. propose computationally efficient methods estimation, simulation, predictio...

Abstract Albeit worryingly underrated in the recent literature on machine learning general (and, deep particular), multivariate density estimation is a fundamental task many applications, at least implicitly, and still an open issue. With few exceptions, neural networks (DNNs) have seldom been applied to estimation, mostly due unsupervised nature of task, (especially) need for constrained train...