Alyssa Andrichik
Eva Licht
Joe Yalowitz
We are investigating the World Happiness Index in order to build a model that can predict a country's Happiness Score based on demographic and geographic factors such as literacy levels, cell-phone use, birth rate, death rate, GDP per capita, perceived corruption, etc. We will build an array of linear models, simple and with interaction, and use other regression analysis tools such as Ridge, Lasso, and PCA to understand our data. We will also use regression trees, random forests, and boosted trees to develop prediction methods for our research question. We have determined by running our training data and validated with our test data, that region is an optimal predictor of happiness.
Our presentation can be found here: https://docs.google.com/presentation/d/1MTjG5kiYoIBgd15PBpHNc3_UT2VoTpzPwrvBq9mjS94/edit?usp=sharing