Selected Projects in Data Science and ML

---

Time Series Forecasting of Stock Volatility and Option Pricing

Fetched market data and conducted risk-return analysis by using various financial metrics on AAPL. Performed EDA, decomposition and tested for stationarity. Analyzed ACF and PACF plots, grid searched best-fitted hyperparameters using parallel programming and built ARMA, ARIMA, and SARIMA models and evaluated their performance. Time-series forecasting on trailing volatility and estimated option pricing with Black-Scholes Model.

Tech: Statsmodels, SciPy, Scikit-learn, NumPy, Pandas, Matplotlib, Seaborn, Python

View code on Colab

---

Malaria Detection with AlexNet and VGG-16

Built effiicient Convolutional Neural Networks (CNN) by using AlexNet and VGG-16 architectures to detect the presence of malaria parasites through the patients’ blood smear images. Performed multiple hyperparameters tuning with Keras, and applied data augmentation and transfer learning.

Tech: TensorFlow, Keras, Scikit-learn, OpenCV, Matplotlib, Seaborn, NumPy, Pandas, Python

View code on Colab

---

Household Finance Dashboard

An interactive PowerBI dashboard which sources data from MySQL database, and performs ETL workflow to update DB with Apache Airflow. The dashboard provides comprehensive and updated financial insights of household to users.

Tech: Apache Airflow, Power BI, MySQL, Pandas, Python

View dashboard

View source code on GitHub

---

Other Projects

• Coding Algorithms from Scratch
• Drug Classification using KNN
• SpaceX Falcon 9 First Stage Landing Prediction

---