When testing for a treatment effect or a difference among groups, the distributional assumptions made about the response variable can have a critical impact on the conclusions drawn. For example, controversy has arisen over transformations of the response (Keene). An alternative approach is to use some member of the family of generalized linear models. However, this raises the issue of selecting the appropriate member, a problem of testing non-nested hypotheses. Standard model selection criteria, such as the Akaike information criterion (AIC), can be used to resolve problems. These procedures for comparing generalized linear models are applied to checking for difference in T4 cell counts between two disease groups. We conclude that appropriate model selection criteria should be specified in the protocol for any study, including clinical trials, in order that optimal inferences can be drawn about treatment differences.

Although models developed directly to describe marginal distributions have become widespread in the analysis of repeated measurements, some of their disadvantages are not well enough known. These include producing profile curves that correspond to no possible individual, possibly showing that a treatment is superior on average when it is poorer for each individual subject, implicitly generating complex and implausible physiological explanations, including underdispersion in subgroups, and sometimes corresponding to no possible probabilistic data generating mechanism. We conclude that such marginal models may sometimes be appropriate for descriptive observational studies, such as sample surveys in epidemiology, but should only be used with great care in causal experimental settings, such as clinical trials.

Methods for the isolation and culture of enriched populations of Sertoli cells from 20-60 day old rats are described. The identity of the Sertoli cells was verified by bright light and electron microscopy. Freshly isolated Sertoli cells specifically bound follicle stimulating hormone (FSH) but not luteinizing hormone (LH) and responded to FSH stimulation with dramatic increase in cyclic AMP level. Isolated Sertoli cells, maintained in culture for 11 days, showed no evidence of proliferation but retained their characteristic ultrastructural features and FSH binding ability. Incubation of cultured cells with FSH resulted in a significant stimulation of cyclic AMP and androgen binding protein (ABP). Since the freshly isolated or cultured cells were predominantly (greater than 80%) Sertoli cells, these results provide direct evidence that the Sertoli cells represent a primary target site for FSH activity in the testes. The culture method also provides a valuable in vitro model for the study of chronic effects of various agents on the Sertoli cell.

SUMMARYWhen testing for a treatment effect or a difference among groups, the distributional assumptions made about the response variable can have a critical impact on the conclusions drawn. For example, controversy has arisen over transformations of the response (Keene). An alternative approach is to use some member of the family of generalized linear models. However, this raises the issue of selecting the appropriate member, a problem of testing non-nested hypotheses. Standard model selection criteria, such as the Akaike information criterion (AIC), can be used to resolve problems. These procedures for comparing generalized linear models are applied to checking for difference in T cell counts between two disease groups. We conclude that appropriate model selection criteria should be specified in the protocol for any study, including clinical trials, in order that optimal inferences can be drawn about treatment differences.

This book was first published in 2004. Many observed phenomena, from the changing health of a patient to values on the stock market, are characterised by quantities that vary over time: stochastic processes are designed to study them. This book introduces practical methods of applying stochastic processes to an audience knowledgeable only in basic statistics. It covers almost all aspects of the subject and presents the theory in an easily accessible form that is highlighted by application to many examples. These examples arise from dozens of areas, from sociology through medicine to engineering. Complementing these are exercise sets making the book suited for introductory courses in stochastic processes. Software (available from www.cambridge.org) is provided for the freely available R system for the reader to apply to all the models presented.