CMStatistics 2015: Start Registration
View Submission - CMStatistics
B0415
Title: Tests for monotonic and nonmonotonic trend in time censored recurrent event data Authors:  Jan Terje Kvaloy - University of Stavanger (Norway) [presenting]
Bo Henry Lindqvist - (Norway)
Abstract: In recurrent event data it is often of interest to detect possible systematic changes in the pattern of events. An example is a repairable system for which it is important to detect changes in the pattern of failures. Another example is to test for changes in the return of a recurrent disease. We say that there is a trend in the pattern of events if the inter-event times tend to alter in some systematic way, which means that the inter-event times are not identically distributed. By using statistical trend tests it is possible to decide whether such an alteration is statistically significant or not. In general a trend in the pattern of events can be either monotonic or nonmonotonic and it is thus useful to have tests with power against both monontonic and nonmonotonic trend. We present a class of statistical tests for trend in the event times in time censored recurrent event data based on the general null hypothesis of a renewal process. This class does in particular include a test which is attractive for general use by having good power properties against both monotonic and nonmonotonic trends.