Analyzing perceived hunger across states in the US

Abstract.

This paper focuses on the problem of analyzing how factors impact hunger across states when hunger is ill-defined. Hunger (which is a latent variable) is presumed to depend on macroeconomic, legislation, policy, and demographic variables. Based on the Bayesian method of a posterior odds ratios, we find that the high school graduation rate appears to be the single most important factor we identify which affects the perceived hunger measure.