project

Research PEARL

At the invitation of Henry Zhao, one of the students at the Welch Center, I am here to present a research PEARL. Henry has been enrolled in both the Stata I & II classes I’ve taught this spring.

The Stata II class has focused on utilizing remote work directories on GitHub, emphasizing:

• Version Control: Ensuring documents are iteratively improved in an organized manner.
• Seamless Collaboration: Facilitating teamwork with both local members at Hopkins and remote collaborators from anywhere.
• Open and Transparent Methods: Creating processes that are easy for others to emulate, critique, and improve.

I believe these reasons contributed to Henry inviting me to this research PEARL. So, without further ado (pun very much intended)…

global repo https://github.com/jhustata/project/raw/main/
do ${repo}welchcenter.do

Stata .do files

• Staight out of Stata!!
• Click here for other .do file sources
• Download Stata_ReadInProgramAllSurveys.do
  • Edit filepath in the .do file
  • Review my edits here for guidance
  • Save edits as followup.do and upload to your username/project repo
  • Any edits should be saved under the same followup.do file name, but use “commit changes” statements for version control
• This script followup.do produces mortality followup data
• Merge these outcomes with NHANES survey (i.e., baseline data)

Stata .ado files

• Long-term plan is to automate the above processes using .ado scripts with programs that allow the unskilled user to:
  • Reproduce the analyses
  • Expand the parameters in the model
  • Consider interaction terms
  • Variable functional forms (e.g. collapsed categories, spline terms, etc)