Food Security in Africa

Food insecurity has resurfaced to be a vital issue in the past two decades, especially in sub-Saharan Africa. The issue is more than just growing more food and feeding people but also delivering sufficient to a national population that are affordable to them. It touches upon issues on poverty, gender equality, water scarcity, climate change, technology adoption, trade, and agricultural policy: practically all aspects of an economy and society.

Things that we care about: income and price shocks; natural disasters such as drought and flood; increased volatility in international and domestic market prices; impacts of agricultural policies.

Current work:

  1. a framework for integrating relevant public data that are different in frequency and geospatial scale

  2. detailed prediction of food security in out-of-sample using machine learning techniques

  3. policy analysis on stockholding

Future work:

  1. build data pipelines to scale and automate the project

  2. prediction of food security at finer geospatial level with help from satelite imageries

  3. more interpretation on model results in the context of early warning

  4. combine machine learning in model selection and heterogenous treatment effect


. Predicting Food Security with Machine Learning. 2019.


. A Data-Driven Approach Improves Food Insecurity Crisis Prediction. World Development, 2019.

PDF Code Project Appendix

. Effects of Stock-holding Policy on Maize Prices: Evidence from Zambia. Journal of Agricultural and Food Industrial Organization, 2019.

PDF Code Project Appendix