Models & Tools

Regression Models

1. Customer Churn Estimation with Loyalty and Novelty Effects (PYTHON)
   • extension to the Hardie & Fader Model to include time-varying churn rates
2. Customer Churn Estimation (PYTHON)
   • predict customer survival (original Hardie & Fader CLTV Model)
3. Scrubbing and Pre-processing Raw Data (PYTHON)
   • preparing raw mortgage data for use in regression analysis
4. Modeling Mortgage Defaults - Part 1: Logistic Regression (R)
   • comprehensive overview of data analysis and model creation process, using real mortgage data
5. Modeling Mortgage Defaults - Part 2: Logistic Transition Models (R)
   • exploration of linked logistic models to capture time-varying transition probabilities
6. Modeling Mortgage Defaults - Part 3: Cox Regression and Hazard Rate Models (R)
   • exploration of a competing risks hazard rate model to predict future portfolio composition / exit attribution

Applied Functions

1. Simple Friend Recommender (PYTHON)
   • predict friend candidates based on proximity and popularity
2. Emerging Market ETF Analysis (R)
   • analysis of Emerging Market ETF returns and optimized allocation % in in US Equity portfolio