This project estimates Markov income transition matrices using Panel Study of Income Dynamics (PSID) data conditional on age and on the key variables from Chetty, et al (2014): high school dropout rate, crime rate (social capital), and inequality (gini coefficient).
We will match geographic variables to the PSID respondents geography at the time of income measurement. So in the end, we will estimate hundreds of transition matrices.