This project focuses on logistic regression testing using 2019 voter party registration data from the state of North Carolina
Date: December 2019
Project Members: Irv Campbell and Steven Dye
Goal: To be able to predict what political party a voter will change their registration to based on what party they were previously registered to and what county they live in.
Responsibilities:
- Define project scope and focus
- Collect data
- Form hypothesis
- Perform exploratory data analysis
- Create Master Notebook
- Create Regression Model
- Test hypothesis
- Create presentation
- Lint/clean code file
Summary of files:
- Master_Notebook.ipynb: Jupyter Notebook documenting the code and the analysis for the project. Written for a technical audience
- Predicting N.C. Voter Party Changes.pdf: PDF of final presentation
- data file
- 2019_party_change_list.csv: 2019 Voter registration data from the state
- X_test.csv: Test features
- X_train.csv: Train features with SMOTE
- y_test.csv: Test target
- y_train.csv: Train target with SMOTE
- data_prep.py: Code used to clean data and to add SMOTE data
- nc_functions.py: Module to store functions
- viz.py: File for storing vizualization functions