KFAS: Exponential Family State Space Models in R

Ph.D. Thesis State Space Models for Exponential Family Data

Bayesian Analysis of A General Class of Multivariate Exponential Family of State Space Models

In this paper, we propose a general class of multivariate exponential family of state space models and consider their Bayesian analysis using particle learning methods. Our proposed model can be considered to be a direct multivariate extension of the class of state space models developed in Gamerman et al. (2013) in the Journal of Time Series Analysis. Unlike most state space time series models...

Prediction Intervals for Exponential Smoothing State Space Models

The main objective of this paper is to provide analytical expressions for forecast variances that can be used in prediction intervals for the exponential smoothing methods. These expressions are based on state space models with a single source of error that underlie the exponential smoothing methods. Three general classes of the state space models are presented. The first class is the standard ...

Forecasting based on state space models for exponential smoothing

In business, there is a frequent need for fully automatic forecasting that takes into account trend, seasonality and other features of the data without need for human intervention. In supply chain management, for example, forecasts of demand are required on a regular basis for very large numbers of time series, so that inventory levels can be planned to provide an acceptable level of service to...

Restructuring exponential family mixture models

Variational KL (varKL) divergence minimization was previously applied to restructuring acoustic models (AMs) using Gaussian mixture models by reducing their size while preserving their accuracy. In this paper, we derive a related varKL for exponential family mixture models (EMMs) and test its accuracy using the weighted local maximum likelihood agglomerative clustering technique. Minimizing var...