A MODEL FOR BINARY TIME-SERIES DATA WITH SERIAL ODDS RATIO PATTERNS

Moment methods for analysing repeated binary responses using the margi nal odds ratio as a measure of association have recently been proposed by several researchers. Using the generalized estimating equation (GE E) methodology, they estimated the regression parameters associated wi th the expected value of an individual's vector of binary responses. I n addition, they estimated the marginal odds ratio between pairs of bi nary responses. In this paper, we discuss a model for binary time seri es data where the repeated responses on each individual may be unequal ly spaced in time. This model allows both the number of observations p er individual and the times of measurement to vary between individuals . Our approach is to model the association between the binary response s using serial odds ratio patterns. This model can be thought of as a binary time series analogue of the exponential correlation pattern so commonly assumed for continuous time series data. Parameter estimates are obtained by using the GEE methodology. The model is illustrated wi th data from an arthritis clinical trial where the response variable i s a binary self-assessment measurement.