Mathematical and statistical analysis of circadian rhythms.

The mathematical and statistical analysis of biological time series is complex and often involves specialised techniques. In this article we review several of these techniques, placing particular emphasis on the usefulness, assumptions and kind of data that they require. Because classical methods of time series analysis often require long spans of data that are not frequently available in biological studies, particularly in clinical circumstances, several alternative techniques are described. Where possible, emphasis is placed upon simple rather than esoteric mathematical descriptions of data. Also covered are problems of interpretation that might arise as a result of the mathematical analysis.