A quantitative analysis of U.S. Treasury yield curves using the Nelson-Siegel model and Principal Component Analysis (PCA).
This project fits and interprets yield curve dynamics, decomposes term structure movements into level, slope, and curvature factors,
and visualizes how interest rate risk evolves over time — built in Python and Jupyter.
A binary classification pipeline to predict loan default risk. Handles class imbalance using SMOTE and benchmarks Logistic Regression,
Random Forest, and Gradient Boosting — with Logistic Regression emerging as the best model by ROC-AUC. Built with scikit-learn and Python.
A classification model that predicts the directional movement of Bitcoin prices using OHLCV features and technical indicators.
XGBoost delivered the strongest discriminating power, with chronological train-test splits preserving temporal integrity. Built with Python and scikit-learn.
An exploratory analysis of historical stock market data using Python. Covers trend identification, return calculations, volatility analysis,
and data visualization using Pandas, Matplotlib, and Seaborn to uncover patterns in equity price movements.
A data analysis project exploring SpaceX Falcon 9 launch records. Applies data wrangling, exploratory analysis, and visualization
techniques to identify patterns in launch success rates, payload mass, and orbital destinations using Python.
A collection of hands-on projects completed as part of DataCamp's Data Analyst certification track. Covers data manipulation,
statistical analysis, and visualization using Python and SQL across real-world datasets.