From January-March 2022, I took a ten-week week course titled Regression Analysis (STA 264), a statistics class at Union College taught by professor Roger Hoerl. The course covered the mathematical framework of regression models, how to perform analysis based on fitted models, how to evaluate data and their sources, and much more.
Throughout the course, we continually developed analysis of a data set of our choosing, which we eventually developed into the final project of the course. Matching my interest in tech for social good, I picked the Women Peace and Security Index, a data set developed by researchers at Georgetown University to quantify the relative peace, security and societal equity of women in 170 countries across the world. My goal for the project was to create a model that predicts rates of intimate partner violence based on the other categories measured by the Women Peace and Security Index with the hope that the resulting regression model would offer insights into the relative impact of phenomena represented by fields in the data set on intimate partner violence.
Throughout the course, we were required to fit our regression models using JMP and were asked to pick a data set with minimal data cleaning needed. Thus, I do not have any code to accompany my model, but I did articulate each of the decisions I made in building my model in my final paper.
Please see the PDF uploaded to this repository for my written analysis of the data set.