Elastic properties of cortical bone in female human femurs.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to U Helfenstein.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
During the last few years Box-Jenkins models have become of increasing importance in such fields as economics and industry. They have been used for forecasting and for detecting relations between different time series. Similar applications are also relevant in epidemiology. Since many people who are concerned with the analysis of medical data are not familiar with this subject, a short non-technical introduction to Box-Jenkins models is given. The model-building process is demonstrated in some detail using monthly case reports of the two seasonal endemic diseases chicken-pox and mumps, and the relation between these two time series is investigated. It is found that both diseases may be represented by simple models which have basically the same statistical structure and that they are related at time-lag 0 but not at other time-lags.
If a patient is treated in a hospital, drugs are administered at 'deterministic', known times. However, if the patient is not supervised the times of drug intake may be less 'deterministic' and less well known. It seems therefore worthwhile to make a theoretical experiment in which one assumes that drugs are applied according to a random process. In the present article it is proposed that shot noise models could be used to represent concentration curves in such situations. By means of Monte Carlo methods different measures of random drug excess are described. An included program may encourage readers to consider these methods and to perform computer experiments.