Space-Time Modeling of Small Area Data in a Developing World Setting

Friday, March 30, 2018 - 3:30pm

Speaker(s): 
Jon Wakefield, University of Washington

Abstract: 

Many people living in low- and middle-income countries are not covered by civil registration and vital statistics systems. Consequently,  household sample surveys with complex designs are often used to estimate health and population indicators. In this talk I will describe two approaches to the space-time modeling of complex survey data.  In a  hybrid model we take as data the weighted estimator of the quantity of interest (along with its variance).  This data model is then combined with a spatio-temporal smoothing model that alleviates problems of data sparsity, which results in a small-area estimation model. An alternative approach includes design effects in the model and uses a continuously-indexed spatial and discrete time model. The effects of modeling the design are investigated through simulation. Issues that will be discussed include:  the simultaneous  use of both point- and area-level data; how to make adjustments for HIV epidemics; the inclusion of so-called indirect data. The modeling of under-5 mortality in Kenya is used to motivate and illustrate the issues raised.  Data come from a variety of sources including Demographic and Health Surveys conducted over the period 2003–2014.

Seminars generally take place in 116 Old Chemistry Building on Fridays from 3:30 - 4:30 pm. For additional information contact: karen.whitesell@duke.edu or phone 919-684-8029. Sorry, but we do not have reprints available. Please feel free to contact the authors by email for follow-up information, articles, etc. Reception following seminar in 211 Old Chemistry

Old Chemistry 116

Location Info