Granular mortality estimation and modeling from sparse data
In this double seminar, the speakers present two methodological innovations in the field of mortality modeling and estimation.
In his presentation, Dr Eugenio Paglino introduces a unified workflow that brings together mortality modeling, life table construction, and uncertainty estimation. Building on the TOPALS approach, he propose a flexible, two-step Generalised Additive Model (GAM) to convert imperfect, sparse, or abridged data into smooth, single-year schedules. By first anchoring empirical data to a standard demographic curve and only then smoothing the localised deviations, the model avoids common collinearity traps. The proposed framework connects classic direct estimation with modern techniques, giving researchers a straightforward way to present both point estimates and uncertainty for any life table quantity.
Indirect demographic estimation methods, first proposed by William Brass (LSHTM) in the late 1960s, rely on simple data inputs (e.g., summary birth histories) and formal demographic relationships to derive estimates of fertility and (under-5) mortality. Given their low data requirements, this class of methods have been widely used in LMICs to estimate demographic rates. José H C Monteiro da Silva revisits indirect methods to convert the proportion of deceased children from summary birth histories into probabilities of dying. The goal of this work is to update the Brass Method for under-5 mortality estimation by integrating it with the Under-5 Log-Quad model to provide more granular estimates of under-5 mortality.
Any externals who wish to attend in person, please report to the LSHTM reception in Keppel Street at 12:30.
Speakers
Dr Eugenio Paglino
Dr Eugenio Paglino is a postdoctoral researcher at the University of Helsinki and he works on migrant health and mortality, climate change and population dynamics, inequalities in health and mortality, and statistical demography.
José HC Monteiro da Silva
José HC Monteiro da Silva is a PhD Candidate in Demography at the University of Pennsylvania and dissertation focuses on Under-5 mortality estimation.
Event notices
- Please note that you can join this event in person or you can join the session remotely
- Please note that this session will not be recorded
Admission
Contact
