In this project, I analyzed survey data from 1605 Pakistanis who participated in Hajj lottery for Hajj visa (an originally analysis was presented in Clingingsmith, Khwaja, and Kremer (2008).

I applied instrumental regression to make causal inferences on the Hajj's effect on people's religiousness, view towards women, psychological well-being, etc., and applied techniques such as Holm’s (Holm 1979) sequentially rejective procedure and Bonferroni style procedures to control type I error from multiple testing. I used post-double-selection with Lasso regression to discover possible covariates in the causal models in a data-driven manner.

The results are described in this paper.