This package provides an implementation of A Bayesian Multivariate Functional Dynamic Linear Model (Kowal, Matteson, Ruppert, JASA, 2017) for modeling a (multivariate) time series of functional data. The model may be described as a dynamic functional factor model: the factor loadings are modeled as smooth curves, while the factors are modeled dynamically as latent state variables in a state space model (or dynamic linear model). The package provides an efficient Gibbs sampling algorithm, as well as functions for component block sampling steps: (1) the factor loading curves, (2) the state variables (factors) using the package KFAS, and (3) the observation and evolution error variances. Modifications of steps (2) and (3) provide a broadly applicable framework for modeling a time series of functional data.
|Author||Daniel R. Kowal <[email protected]>|
|Maintainer||Daniel R. Kowal <[email protected]>|
|Package repository||View on GitHub|
Install the latest version of this package by entering the following in R:
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.