CMStatistics 2018: Start Registration
View Submission - CMStatistics
B0835
Title: Goodness-of-fit test for compound Cox process Authors:  Paula Bouzas - University of Granada (Spain) [presenting]
Nuria Ruiz-Fuentes - University of Jaen (Spain)
Abstract: A goodness-of-fit test is proposed for a compound Cox process when it is observed in a discrete set of time points; moreover, the available information can be collected as recurrent event data or panel count data. The test determines if a new sample path fits a given model whether it is already known or it has to be estimated previously. In the latter case, functional data analysis is used throughout the procedure. The assessment of the goodness-of-fit test is carried out by means of its application to several simulation cases. In each case, a number of new sample paths with or without provoked perturbations are tested to prove the accuracy of the proposed hypothesis test.