Model-Based Estimates in Demography and Global Health: Quantifying the Contribution of Population-Period-Specific Information
vendredi 17 juin 2022, 11h00 - 12h00
Salle du conseil, espace Turing
Sophisticated statistical models are used to produce estimates for demographic and health indicators even when data are limited, very uncertain or lacking. For example, country-specific estimates of unintended pregnancies and abortions from 2000 to 2020 are often based on information regarding family planning, births and intention status of births only. Estimation is carried out based on a fertility accounting model, with group-specific pregnancy and abortion rates modeled with Bayesian hierarchical time series models (Bearak et al., 2020).
To facilitate interpretation and use of model-based estimates, we aim to provide a standardized approach to answer the question: To what extent is a model-based estimate of an indicator of interest informed by data for the relevant population-period as opposed to information supplied by other periods and populations and model assumptions? We propose a data weight measure to calculate the weight associated with population-period data set y relative to the model-based prior estimate obtained by fitting the model to all data excluding y. In addition, we propose a data-model accordance measure which quantifies how extreme the population-period data are relative to the prior model-based prediction.
In this talk, I will first introduce the fertility accounting model. The second part of the talk focuses on the new data weight and data-model accordance measures. I will illustrate the insights obtained from the combination of both measures in toy examples and present preliminary selected findings for estimates of unintended pregnancy and abortion rates. This is joint work with Guandong (Elliot) Yang and Krista Gile.
J. Bearak, A. Popinchalk, B. Ganatra, A. Moller, Ö. Tunçalp, C. Beavin, L. Kwok, L. Alkema (2020). Unintended Pregnancy and Abortion by Income, Region, and the Legal Status of Abortion: Estimates from a Comprehensive Model for 1990-2019. The Lancet Global Health 8(9): e1152-61.