Timothy Miller

Overview

Timothy Miller is a demographer at the Center for the Economics and Demography of Aging (CEDA), University of California, Berkeley. His work focuses on stochastic demographic forecasting for Social Security, including mortality projections, stochastic immigration modeling, and evaluating the Lee-Carter forecasting method.

Key Contributions / Features

• Evaluating the Lee-Carter mortality method (Lee and Miller 2001, Demography): Co-authored the benchmark evaluation study showing that a weighted SVD implementation of the Lee-Carter model substantially improves out-of-sample performance, making it the standard Lee-Carter implementation for subsequent SSA forecasting work.
• Stochastic immigration forecasting (Lee, Miller, and Anderson 2004, Report I): Lead author of the stochastic immigration forecast showing that adding probabilistic immigration makes negligible difference to the old-age dependency ratio uncertainty; preferred forecast drops net immigration to ~1M by 2020 then rises slowly to 1.2M by 2100.
• Stochastic infinite horizon SS forecasts (Lee, Miller, and Anderson 2004): Co-authored three-report NBER working paper extending the Lee-Tuljapurkar framework to stochastic immigration, parameter uncertainty, and 500-year infinite horizon actuarial balance (−5.15% of payroll).

Related

• Ronald Lee
• Michael Anderson
• Long-Term Actuarial Balance
• Lee-Carter Model
• Probabilistic Population Forecasting
• SSA Mortality Forecasting

Sources

• Lee, Miller, and Anderson 2004 — Stochastic Infinite Horizon Forecasts for Social Security and Related Studies
• Booth Tickle and Smith 2005 — Evaluation of the Variants of the Lee-Carter Method of Forecasting Mortality A Multi-Country Comparison