This script compiles data from multiple excel files to create 3 separate regression models using cultural indices defined by Professor Geert Hofstede to explain national happiness levels from the 2018 world happiness report. The first model is a regression using just the cultural variables to explain happiness, while the later two models add features to limit omitted variable bias. Finally, the script includes several functions that analyze how statistical significance of the cultural indices changes between models to explain which omitted variables were responsible for the biased estimates and how they were responsible for such errors.
Effectively, the analysis seeks to understand what aspects of national culture are responsible for the level of happiness of that country, and points out through what mechanisms these factors have effect.