Yaşam Çözümlemesinde Aykırı Değerler

Abstract

Survival analysis is a collection of statistical methods for analyzing data where the outcome variable is the time until the occurrence of an event of interest. Outliers are the individuals whose survival time is not well fitted by the model. Outliers in survival anaysis calculated differently from classical regression analysis. Because in survival analysis are data contain censored observations and this makes difficult to calculate the difference "observed minus predicted" values.In survival analysis outlier detection methods are commonly carried out based on residuals and residual analysis. In survival analysis there are different types of residuals that are Cox-Snell, Martingale, Schoenfeld, Deviance, Log-odds and Normal deviance residuals. There are methods which are DFBETA, LMAX and Likelihood Displacement values for detecting influential observations.The residuals are analyzed during the study which is applied on a stomach cancer data set and the outliers are detected.