| dc.description.abstract | Survival data analysis is concerned with the study of time until the occurrence of an event of interest, such as the death of a patient, the cure or the recurrence of a disease. If the exact time of the event of interest is not known, but instead, the event is known to have occurred during aparticular interval of time, the data are known as interval-censored survival data. However, if the diagnostic tool used to detect failure is not perfectly sensitive and specic, the subjects may be misclassi ed: a healthy one may be diagnosed as sick and a sick individual may be diagnosed as healthy as well. In such cases, the traditional survival analysis methods produce biased estimates for the failure time distribution parameters [Paggiaro and Torelli (2004)]. So we developed a model that incorporates sensitivity and specicity in grouped survival data analysis (a special case of minterval- censored data in which all subjects are tested at predetermined time points). Monte Carlosimulation studies have shown that, if sensitivity and specificity are known, the proposed method is very eficient, since its estimates percent bias are lower than those provided by traditional method. | |
