M. C. Pardo Llorente, R. Alonso Sanz

Preisser and Qaqish (1996) proposed deletion diagnostics for GEEs. This approach is a global influence analysis which assesses the individual influence of subjects or observations on the estimates of regression parameters. A subject may have influence on the regression coefficients, the variance of the estimated coefficients, the fitted values, and/or the goodness-of-fit statistics. The global influence diagnostics based on the case-deletion had been generalized from linear regression to GEEs to measure the impact of an observation on all above regression outputs except on the variance of the estimated coefficients. Pardo \& Alonso (2012) proposed global influence measures based on the volume of confidence ellipsoids for detecting influence of subjects on the efficiency of coefficient estimations in GEEs. We define a family of influence measures based on the family of divergence measures of Rényi which extends one measure of this paper. Some simulation studies have been done.

Palabras clave: generalized estimating equations, influence

MD3 Bioestadística 2
17 de abril de 2012  15:30
Sala Londres

A. Martín Andrés, M. Álvarez Hernández

A. E. Marín Jiménez, J. A. Roldán Nofuentes, J. D. Luna del Castillo