Influence diagnostics in generalized log-gamma regression models

摘要

We discuss in this paper application of influence diagnostics in generalized log-gamma regression models considering the possibility of censored observations. WE derive the appropriate matrices for assessing the local influence on the parameter estimates as well as the predictions from the fitted model under different perturbation schemes. The effect of censoring on local influence is also investigated. An example, for which we apply the diagnostic methods, is given as illustration.