Hazard Estimation with Flexible Tails

Polynomial splines are used to estimate the log-hazard function based on possibly censored, positive data. Two additional log terms are incorporated into the fitted model for the log-hazard function to allow for greater flexibility in the extreme tails. A fully automatic procedure involving the maximum likelihood method, stepwise knot addition, stepwise knot deletion and BIC is used to select the final model. A user interface based on S is described for obtaining estimates of the hazard function, density function, distribution function and quantile function and for generating a random sample from the estimated distribution.