|
Title
|
|
Recurrent-Events Analysis for Reliability, Biomedical, Sociological, and Other Applications
|
|
Date / Time / Room
|
Sponsor
|
Type
|
08/13/2002
8:15 AM -
4:15 PM
Room: H-Clinton Suite
|
ASA
|
Other
|
|
Organizer:
|
n/a
|
|
Chair:
|
n/a
|
|
CE Presenter
|
- Wayne Nelson Statistical Consulting
|
Description
This course provides a practical introduction to analyses of repeated events data. Such analyses were recently developed and are widely useful. Survival data typically consist of a single event (end of life) for each population unit, which is modeled with a life distribution. In contrast, repeated-events data can have any number and times of events for a unit. Course examples include repairs of products, recurrences of bladder tumors in patients, consumer purchases on the Internet, and births of children to statisticians. Such data are modeled here with a simple, versatile nonparametric stochastic process.
While emphasizing reliability applications (product repair date), the course includes many applications from other fields. The course shows how to make and interpret informative data with confidence limits for common types of recurrence data, which are censored. These plots are as basic and as informative as are probability plots for univariate survival and other data. The course also provides such plots for "costs" and other numerical "values" of recurrences. In contrast, most books and literature deal only with counts of recurrences. Finally, the course briefly surveys basic parametric models (homogeneous and non-homogeneous Poisson processes and renewal processes) and regression models (Poisson and Cox ) for such data.
While presenting new concepts, this course requires only a background in basic statistics. Background in stochastic processes is unnecessary. Participants will receive the final manuscript of Wayne Nelson's new book RECURRENT-EVENTS DATA ANALYSIS FOR PRODUCT REPAIRS, DISEASE RECURRENCES, AND OTHER APPLICATIONS, which will be published by ASA-SIAM in September 2002 upon which the course is based.
|
|
