// Project Portfolio
MSCSIS Graduate
Building quantitative tools in market risk, credit risk, and portfolio stress testing, using Python, live market data, and industry-standard methodologies.
// Portfolio at a glance
Python engine implementing Historical, Parametric, and Monte Carlo VaR with rolling backtesting, Kupiec POF statistical validation, and visual diagnostics for portfolio tail risk.
Surface-weighted Elo model trained on 3 years of ATP data. Simulates the 2024 AO bracket 100,000 times to estimate title probabilities — Jannik Sinner ranked 2nd at 16.1%.
R-based time series analysis forecasting sunspot activity, solar flux, and geomagnetic disturbances using ARIMA/SARIMA, Vector ARIMA, and dynamic regression on NOAA space weather data.
Databricks medallion ETL pipeline ingesting live rental listings via RentCast API, transforming through Bronze/Silver/Gold layers, and using the Claude AI API to score each listing against preferences.
// Who I am
Background
Focused on building practical risk analytics skills to enter the finance sector. Self-directed learner working through market risk, credit risk, and quantitative methods, building every concept into a working tool rather than just studying theory.
Currently learning
Deepening knowledge in quantitative risk methods through self-directed study and hands-on project work, building every concept into a working tool rather than studying theory alone.
Technical skills
Python · pandas · NumPy · SciPy · scikit-learn · SQL · Git · yfinance
Risk domains
Market risk · Credit risk · Operational risk · Stress testing · VaR · CVaR · Expected Loss · Basel III frameworks