Logo Lanfrica

bihebe90/socioeconomic-portfolio

Domain:

socioeconomic
Creator:
bih
Host:
Analyse de la vulnérabilité des ménages au Togo - Portfolio R # Socioeconomic Vulnerability Analysis – Togo ## Objective This project analyzes household vulnerability in Togo using survey data and statistical modeling in R. ## Methodology - Data cleaning and preprocessing - Construction of a vulnerability index (water, sanitation, energy) - Descriptive analysis - Logistic regression model - Model evaluation (ROC, AUC) ## Key Results - 25.9% of households are classified as vulnerable, based on a multidimensional threshold (at least 2 deprivations in water, sanitation, and energy) - Rural households are significantly more likely to be vulnerable than urban households - Strong regional disparities exist, with Plateaux and Kara showing higher vulnerability levels - The logistic model shows moderate predictive performance (AUC ≈ 0.65), suggesting that additional socioeconomic variables are needed ## Limitations The model is based on limited variables and does not include income, education, or housing conditions. ## Reproducibility ```r source("run_all.R") ```