|
Activity Number:
|
380
- Curious Roles of Latent Variables in Prediction and Inference
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, July 30, 2019 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Mental Health Statistics Section
|
|
Abstract #300588
|
|
|
Title:
|
Forecasting Future Smoking-Related Mortality in 69 Countries: The Vital Role of Latent Variables
|
|
Author(s):
|
Yicheng Li and Adrian Raftery*
|
|
Companies:
|
University of Washington and University of Washington
|
|
Keywords:
|
Life expectancy;
Bayesian hierarchical model;
forecast interval;
latent variable;
smoking;
mortality
|
|
Abstract:
|
Smoking is one of the main preventable threats to human health. Estimating and forecasting the smoking attributable fraction (SAF) of mortality can yield insights into smoking epidemics and also provide a basis for more accurate mortality and life expectancy projection. Peto et al (1992) proposed a method to estimate the SAF using the lung cancer mortality rate as an indicator of exposure to smoking in the population of interest. Here we use the same method to estimate the SAF for both sexes for 69 countries for 1950-2015. We document a strong and cross-nationally consistent pattern of the evolution of the SAF over time. We use this as the basis for a Bayesian hierarchical model to project future male and female SAF from 69 countries simultaneously. This gives forecasts as well as predictive distributions that can be used to find uncertainty intervals for any quantities of interest. We assess the model using out-of-sample predictive validation, and find that it provides good forecasts and well calibrated forecast intervals. The time evolution of SAF for each country is viewed as a latent variable, and this is vital to the success of the method.
|
Authors who are presenting talks have a * after their name.
