摘要：For many years actuaries and demographers have been doing curve tting of age-speci c mortality data. We use the eight-parameter Heligman-Pollard (HP) empirical law to t the mortality curve. It consists of threenonlinear curves, child mortality, mid-life mortality and adult mortality. Itis now well-known that the eight unknown parameters in the HP law aredicult to estimate because numerical algorithms generally do not convergewhen model tting is done. We consider a novel idea to t the three curves(nonlinear splines) separately, and then connect them smoothly at the twoknots. To connect the curves smoothly, we express uncertainty about theknots because these curves do not have turning points. We have importantprior information about the location of the knots, and this helps in the estimationconvergence problem. Thus, the Bayesian paradigm is particularlyattractive. We show the theory, method and application of our approach.We discuss estimation of the curve for English and Welsh mortality data.We also make comparisons with the recent Bayesian method.
关键词：Beta-binomial model; Gibbs sampler; median life; over parameterization;splines.
