The data used in this project can be found at https://github.com/ecollins/TUP-data
It's kept separately since 1) it is seldom modified, 2) contains large files, and 3) is used by another separate project.
Contains
-
neediness.org, which contains the code and text of the paper. Most of the python files in the analysis folder are tangled from neediness.org.
-
HighFrequency_Consumption.org, which contains the code to wrangle the mobile survey in the data repository.
Contains
-
TUP.py, which contains the central functions for reading the TUP survey data, getting it into useable formats for asset and consumption analysis, running regressions using the statsmodels package, and converting dataframes to org tables (including regression tables with standard errors and stars).
-
estimation, a folder including essential functions for estimating marginal utilities and relating them back to the Frisch demand model.