Count time series models
June 4, 2015 — December 9, 2015
Statistical models for time series with discrete time index and discrete state index, i.e. lists of non-negative whole numbers with a causal ordering.
C&c symbolic dynamics, nonlinear time series wizardry, random fields, branching processes and Galton Watson processes for some important special cases. If there is no serial dependence, you might want unadorned count models.
1 Maximum processes
Series monotonically increasing at a decreasing rate? Perhaps you have a maximum process.
2 Finite state Markov chains
Often fit as if non-parametric, although there exist parametric transition tables if you’d like, and if you have a large state space you probably would like.
3 GLM-type autoregressive
GLMs applied to time series. Fokianos et al.
4 Linear branching-type and self-decomposable
A.k.a. INAR(p), GINAR(p), INMA.
See Galton Watson processes and other branching processes.
5 Queueing models
See Queueing models.
6 Other
Non-linear processes, arbitrary interaction dynamics, Turing machines, discretised continuous processes…
Todo: get a handle on Twitter’s Robust anomaly detection
This paper proposes a simple new model for stationary time series of integer counts. Previous work has focused on thinning methods and classical time series autoregressive moving-average difference equations; in contrast, our methods use a renewal process to generate a correlated sequence of Bernoulli trials. By superpositioning independent copies of such processes, stationary series with binomial, Poisson, geometric or any other discrete marginal distribution can be readily constructed. The model class proposed is parsimonious, non-Markov and readily generates series with either short- or long-memory autocovariances.