Stock Market Analysis Using Machine Learning and Artificial Intelligence
Trading Indicator Importance
Used 5 years worth of training data to run a machine learning algorithm that would help decide which indicators are most important.
Using XGBoost Regressor
After deciding what features to use, I used XGBoost Regressor to train the model and predict what the daily closing price of the S&P 500 would be.
LSTM Neural Network
To compare results, I trained a long short term memory neural network to predict the daily close of the S&P 500 based off of the previous days open, high, low and close.
Goal
The goal for this project was to learn more about what indicators play a significant role in predicting the closing price of the S&P500 and to be able to forecast future potential opening prices using Long Short Term Memory Neural Networks.
Experienced Gained
- Forecasting models using AI and machine learning algorithms.
- TensorFlow and Sci-Kit Learn.
- Greater understanding of Python and data analysis with large datasets.
Food Safety in Sub-Saharan Africa
MCBAC 2023
Analyzed food security in Sub-Saharan Africa for 2023 Manhattan College of Business Analytics Competition.
Optimal Cold Storage Units
Performed a K-Means clustering algorithm to find the optimal space between countries in Africa.
Competition Presentations
Spent three days at Manhattan College of Business competing against other top schools.
Goal
I was selected along four others to participate on a data analytics competition team in the 2023 spring semester. Our task was to analyze hundreds of thousands of rows of data to propose a solution to food safety and security in Sub Sahara Africa.
Experienced Gained
- Team collaboration and problem solving.
- Analyzing large datasets to form a business decision.
- Greater understanding using machine learning to create target markets.