LASSO for Public Health Data: An Examination of Prevalent Variable Selection Methods and Demonstration of LASSO in R

This repository holds all of the files necessary to recreate Suzanne Dufault's 2017 MA thesis for completion of the MA in Biostatistics at the University of California, Berkeley.

It contains the following subdirectories:

• reports contains all of the .Rnw files
• docs contains all of the knited .pdf files
• graphs contains any generated graphs

The data is publicly available with permission from Young Lives. Feel free to contact me if you would like to know which variables in particular where used from this dataset for the analysis.

To run this on your own computer, you will need to comment out the Tutorial section, which requires access to the YoungLives dataset.