A Bayesian Adjustment of the HP Mortality Law Using a Nonlinear Switching Regression Model

• Activity Number: 168
• Type: Contributed
• Date/Time: Monday, August 1, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Dilli Bhatta*+ and Balgobin Nandram and Rong Wei
• Companies: Worcester Polytechnic Institute and Worcester Polytechnic Institute and Centers for Disease Control and Prevention
• Keywords: Mortality ; Helligman-Pollard ; Switching Regression ; Bayesian

Abstract

Curve fitting of mortality for the US population is an important project at the National Center for Health Statistics (NCHS). We use the eight-parameter Heligman-Pollard (HP) empirical law to fit the curve. It consists of three nonlinear curves: child mortality curve, mid-life mortality curve and adult mortality curve added together. It is now well-known that the eight unknown parameters in the HP law are difficult to estimate. We consider a novel idea to fit the three curves (splines) separately, and then connect them smoothly at the two knots. Because these curves do not have turning points, to connect the curves smoothly we express uncertainty about the knots. A priori we have ranges of values for the knots. Thus, the Bayesian paradigm is particularly attractive. We discuss estimation of the mortality curves for US data, 1999-2001, and English and Welsh data, 1988-1992.