Demographic forecasting

Demographic Forecasting introduces new statistical tools that can greatly improve forecasts of population death rates. Mortality forecasting is used in a wide variety of academic fields, and for policymaking in global health, social security and retirement planning, and other areas. Federico Girosi and Gary King provide an innovative framework for forecasting age-sex-country-cause-specific variables that makes it possible to incorporate more information than standard approaches. These new methods more generally make it possible to include different explanatory variables in a time-series regression for each cross section while still borrowing strength from one regression to improve the estimation of all. The authors show that many existing Bayesian models with explanatory variables use prior densities that incorrectly formalize prior knowledge, and they show how to avoid these problems. They also explain how to incorporate a great deal of demographic knowledge into models with many fewer adjustable parameters than classic Bayesian approaches, and develop models with Bayesian priors in the presence of partial prior ignorance. By showing how to include more information in statistical models, Demographic Forecasting carries broad statistical implications for social scientists, statisticians, demographers, public-health experts, policymakers, and industry analysts. Introduces methods to improve forecasts of mortality rates and similar variables Provides innovative tools for more effective statistical modeling Makes available free open-source software and replication data Includes full-color graphics, a complete glossary of symbols, a self-contained math refresher, and more "Demographic Forecasting introduces new statistical tools that can greatly improve forecasts of population death rates. Mortality forecasting is used in a wide variety of academic fields, and for policymaking in global health, social security and retirement planning, and other areas. Federico Girosi and Gary King provide an innovative framework for forecasting age-sex-country-cause-specific variables that makes it possible to incorporate more information than standard approaches. These new methods more generally make it possible to include different explanatory variables in a time-series regression for each cross section while still borrowing strength from one regression to improve the estimation of all. The authors show that many existing Bayesian models with explanatory variables use prior densities that incorrectly formalize prior knowledge, and they show how to avoid these problems. They also explain how to incorporate a great deal of demographic knowledge into models with many fewer adjustable parameters than classic Bayesian approaches, and develop models with Bayesian priors in the presence of partial prior ignorance. By showing how to include more information in statistical models, Demographic Forecasting carries broad statistical implications for social scientists, statisticians, demographers, public-health experts, policymakers, and industry analysts."-- Publisher's description Qualitative View -- Existing Methods For Forecasting Mortality. Methods Without Covariates ; Methods With Covariates -- Statistical Modeling. The Model ; Priors Over Grouped Continuous Variables ; Model Selection ; Adding Priors Over Time And Space ; Comparisons And Extensions -- Estimation. Markov Chain Monte Carlo Estimation ; Fast Estimation Without Markov Chains -- Empirical Evidence. Illustrative Analyses ; Comparative Analyses ; Concluding Remarks ; A. Notation -- B. Mathematical Refresher -- C. Improper Normal Priors -- D. Discretization Of The Derivative Operator -- E. Smoothness Over Graphs. Federico Girosi And Gary King ; With Contributions From Kevin Quinn And Gregory Wawro. Includes Bibliographical References (p. [251]-257) And Index. 9780691130941......Page 1 prelims......Page 2 chapter1......Page 22 chapter2......Page 40 chapter3......Page 64 chapter4......Page 76 chapter5......Page 95 chapter6......Page 115 chapter7......Page 145 chapter8......Page 166 chapter9......Page 180 chapter10......Page 191 chapter11......Page 196 chapter12......Page 217 chapter13......Page 232 appendixa......Page 234 appendixb......Page 240 appendixc......Page 262 appendixd......Page 268 appendixe......Page 270 bibliography......Page 272 index......Page 280