Project for CS 6190 Fall 2019. Predicting direction of S&P 500 moves from previous day market data.
Files:
GetStockData.ipynb is a Jupyter notebook that collects 20 years of market data from specified indecies and saves data to spreadsheet.
DataProcess.m is a short matlab script that processes data into that normalizes feature weight data and splits it into training and testing datasests
CSV files are either processed market data (IndexChange.csv) or processed file outputs from DataProcess.m
StocksNeuralNet.ipynb contains the code for the neural net logistic regression of the stock data
MCMCStocks.ipynb contains the code for the MCMC logistic regression of the stock data
To Run:
Neural Network: Download training and test data files and run StocksNeuralNet.ipynb
MCMC: Download traing and test data files and run MCMCStocks.ipynb