Modelling time series of counts with overdispersion

The time series of counts observed in practice often exhibit overdispersion. The INGARCH(p, q) models are able to describe integer-valued processes with overdispersion. Known properties of these models, however, are nearly exclusively restricted to the special case p = q = 1. In this article, we derive a set of equations from which the variance and the autocorrelation function of the general case can be obtained. We investigate the purely autoregressive INGARCH(p, 0) models and show that they are closely related to the standard AR(p) models. For p = 1, we determine the marginal distribution in terms of its cumulants. A real-data example highlights potential fields of application of the INGARCH(p, 0) models.     

Monitoring correlated processes with binomial marginals

Few approaches for monitoring autocorrelated attribute data have been proposed in the literature. If the marginal process distribution is binomial, then the binomial AR(1) model as a realistic and well-interpretable process model may be adequate. Based on known and newly derived statistical properties of this model, we shall develop approaches to monitor a binomial AR(1) process, and investigate their performance in a simulation study. A case study demonstrates the applicability of the binomial AR(1) model and of the proposed control charts to problems from statistical process control.     

Nearest neighbor hazard estimation with left-truncated duration data

Duration data often suffer from both left-truncation and right-censoring. We show how both deficiencies can be overcome at the same time when estimating the hazard rate nonparametrically by kernel smoothing with the nearest-neighbor bandwidth. Smoothing Turnbull’s estimator of the cumulative hazard rate, we derive strong uniform consistency of the estimate from Hoeffding’s inequality, applied to a generalized empirical distribution function. We also apply our estimator to rating transitions of corporate loans in Germany.     

An algorithm for discriminating populations with unequal variance matrices

A geometrical approach is considered to the problem of discriminating between any two multivariate normal populations, in particular, those with unequal variance matrices. This approach guarantees the existence of a linear discriminant function and provides a simple algorithm to derive the equation of this line. A numerical example is provided.     

Rule generation for categorical time series with Markov assumptions

Several procedures of sequential pattern analysis are designed to detect frequently occurring patterns in a single categorical time series (episode mining). Based on these frequent patterns, rules are generated and evaluated, for example, in terms of their confidence. The confidence value is commonly interpreted as an estimate of a conditional probability, so some kind of stochastic model has to be assumed. The model is identified as a variable length Markov model. With this assumption, the usual confidences are maximum likelihood estimates of the transition probabilities of the Markov model. We discuss possibilities of how to efficiently fit an appropriate model to the data. Based on this model, rules are formulated. It is demonstrated that this new approach generates noticeably less and more reliable rules.