The validity conditions for univariate or multivariate analyses of repeated measures are highly sensitive to the usual assumptions. In cancer experiments, the data are frequently heteroscedastic and strongly correlated with time, and standard analyses do not perform well. Alternative...

When testing the equality of two population frequencies, one well-known and common situation is that the test is based on two independent samples. In this paper, we consider the other interesting case, in which the comparison is actually within a single population and the test is based on a...

This paper discusses the problem of designing a new variable sampling plan. Suppose that the lot quality characteristic obeys an exponential distribution. Adopting Taguchi's loss function, the objective is to design a plan under which the producer's risk of rejecting a lot that has a specified...

Bayesian dynamic linear models (DLMs) are useful in time series modelling, because of the flexibility that they off er for obtaining a good forecast. They are based on a decomposition of the relevant factors which explain the behaviour of the series through a series of state parameters....

In this paper, we develop a new approach for modelling underreported Poisson counts. The parameters of the model are estimated by Markov chain Monte Carlo simulation. An application to a real data set from a Portuguese marketing survey illustrates the fruitfulness of the approach. We find that...

An empirical Bayes (EB) model to estimate the exchange rate of a national currency is described. The national currency under consideration is typically non-convertible, and is generally associated with a weak economy of a Third World country. We take the Nigerian currency as an example. Using...

In this paper, a new sampling scheme called the 'conditional double sampling scheme' (CDSS) has been proposed. A compact table is presented for the selection of a CDSS indexed by various combinations of entry parameters. Advantages of the CDSS over the single sampling scheme have been discussed....

Logistic regression is a popular method of relating a binary response to one or more potential covariables or risk factors. In 1980, Hosmer and Lemeshow proposed a method for assessing the goodness of fit of logistic regression models. This test is based on a chi-squared statistic that compares...

Least-squares regression is not appropriate when the response variable is circular, and can lead to erroneous results. The reason for this is that the squared difference is not an appropriate measure of distance on the circle. In this paper, a circular analog to least-squares regression is...

In this paper, we present a Bayesian analysis of exponential power mixture models in the presence of a covariate. Considering Gibbs sampling with MetropolisHastings algorithms, we obtain Monte Carlo estimates for the posterior quantities of interest.