Skip to content

Latest commit

 

History

History
20 lines (12 loc) · 2.32 KB

README.md

File metadata and controls

20 lines (12 loc) · 2.32 KB

Application of Bayesian spatial models for producing ADM2 level estimates of inadequate micronutrient intake using Household Consumption and Expenditure Surveys

The Modelling and Mapping Inadequate Micronutrient Intake (MIMI) team at WFP estimates and predicts indicators of inadequate micronutrient intake across multiple Low- and Middle-Income Countries (LMICs). Geographic disaggregation and mapping of these estimates can be used by nutrition policy and programme decision-makers to identify high risk areas and assess the equity of planned nutrition interventions such as Large-Scale Food Fortification (LSFF). The MIMI team ensures that this evidence is provided to stakeholders through direct engagement and through platforms such as WFP’s HungerMapLive.

Where available, the MIMI team uses Household Consumption and Expenditure Surveys (HCES) to estimate the risk of inadequate micronutrient intake. However, as most HCES are only representative to the first administrative level (ADM1), estimates at finer spatial resolutions are unstable due to data sparsity.

This project explores whether stable estimates of inadequate micronutrient intake at the ADM2 level can be produced through the application of Bayesian spatial modelling techniques.

This case study will use pre-processed household level nutritional adequacy data that has been prepared by the MIMI team using data from the Nigeria Living Standards Survey 2018-19: https://microdata.worldbank.org/index.php/catalog/3827

How to use this repository

To be given access to the data required to run the scripts in this repository, or if you would like to contribute, please contact us at: HQ.MIMIGitHub@wfp.org

The repository currently contains the following R scripts that perform the following functions:

  • src/01apparent_intake_functions.R - This script defines functions that will be required to compile the pre-processed MIMI data to calculate apparent household micronutrient intake.
  • src/02indicator_preprocessing.R - This script produces binary household micronutrient intake inadequacy variables.
  • src/03adm2_estimates.R - This script estimates and maps prevalence of inadequate micronutrient intake aggregated at the ADM2 (Local Government Area) level. A BYM2 model is implemented to produce smoothed prevalence estimates.

Further work to follow...