ABSTRACT
Bayesian Demographic Estimation and Forecasting presents a new approach to demography that combines ideas from a branch of statistics known as Bayesian statistics with ideas from mathematical demography. We illustrate our approach with an example of estimating mortality rates for a small population, with randomly-rounded data.
The methods in the book focus on groups of people, as defined by characteristics such as age, sex, and region, rather than on individuals. We make a clear distinction between (i) the true quantities that we wish to infer, and (ii) the data on which our inferences are based. We also distinguish between (i) the underlying risks or propensities, and (ii) the random events governed by these risks or propensities. The methods in the book emphasize disaggregation, forecasting, unreliable data, and demographic systems. A demographic system is a set of interrelated demographic series, such as births, deaths and migration. We present diagrams for three general frameworks, aimed at progressively more complicated applications.
