Code for the final project of the course "Introduction to Causal Inference", entitled "Entrepreneurship in Uganda".
It has been provided in two formats; an RStudio project and an interactive Jupyter notebook.
If you have the conda package manager, e.g. via Miniconda or Anaconda, you can create a virtual environment with all of the dependencies for the code installed, by running in the terminal (with pwd Causal_Uganda):
conda env create -f virtual_environment.yaml
Then, to activate this virtual environment, input into the terminal:
conda activate causal_uganda
Now you have all of the dependencies for this project installed. To run the code, do one of the following:
jupyter notebook ./report/Notebook.ipynb
conda install RStudio
RStudio
Then, from within the RStudio GUI, open the file Causal_1_Uganda.Rproj.
This is available as an Excel Spreadsheet in this directory: SEED_balance_table.xlsx. It is also viewable as a Jupyter notebook, data_cleaning.ipynb, which also gives the code importing and converting the Excel spreadsheet. This was what was used, with some modifications, to create the version presented in the report.
The files for creating the report are available in the report directory. That directory includes the Jupyter notebook with the code for the analysis, the causal diagram produced by the R script Uganda_DAG.R in the main directory, and all of the LaTeX files used to compile the report document. The master LaTeX file is ./report/LaTeX-files/main/report.tex. Compiling that file will reproduce the entire report. The final result report.pdf is also included in the root directory.
This is presentation.pdf.