Gaussian Graphical Models for Partially Observed Multivariate Functional Data
By: Marco Borriero , Luigi Augugliaro , Gianluca Sottile and more
Potential Business Impact:
Finds hidden patterns in messy, incomplete data.
In many applications, the variables that characterize a stochastic system are measured along a second dimension, such as time. This results in multivariate functional data and the interest is in describing the statistical dependences among these variables. It is often the case that the functional data are only partially observed. This creates additional challenges to statistical inference, since the functional principal component scores, which capture all the information from these data, cannot be computed. Under an assumption of Gaussianity and of partial separability of the covariance operator, we develop an EM-type algorithm for penalized inference of a functional graphical model from multivariate functional data which are only partially observed. A simulation study and an illustration on German electricity market data show the potential of the proposed method.
Similar Papers
Conditionally specified graphical modeling of stationary multivariate time series
Statistics Theory
Finds hidden connections in changing data.
Conditionally specified graphical modeling of stationary multivariate time series
Statistics Theory
Finds hidden connections in changing data.
A novel generalized additive scalar-on-function regression model for partially observed multidimensional functional data: An application to air quality classification
Methodology
Helps computers understand messy data with missing parts.