Title: Predicting Short-Term Stock Movements with Quantitative Finance and Machine Learning in Python
Author: Divya Arora (Reg. No. 1901423)
University: University of Essex, Dept. of Mathematical Sciences
Programme: MSc — MA981-7-FY
Supervisor: Dr. Andrew Harrison
Submitted: September 2020
This dissertation identifies the best model for predicting stock market values by comparing traditional quantitative finance methods with modern machine learning approaches — including Random Forest, SVM, ARIMA, and LSTM Recurrent Neural Networks — implemented in Python.
Python · NumPy · Pandas · Scikit-learn · TensorFlow · Keras
- Quantitative Finance (Naive, Average, Moving Average)
- Conventional ML: ARIMA, SVM, Random Forest
- Deep Learning: LSTM RNN
- Data preprocessing, EDA, model evaluation