Predicting the Stock Price Movements Using Indicators

Project description

In this project, a number of indicators are defined based on the minimum, maximum and closing price of public companies for a given period of days. The goal is to predict the direction of price movement, hence, it is a binary classification problem.

Data scources:

Purchased data source that provides the pricing data of public companies.

Feature Engineering : Defining indicators using minimum, maximum and closing stock price as well as their volatility.

Modeling: This project falls into the category of time series. Neural network is used to model the relation between indictaor based features and historical stock price movements.

For further details, please contact me at: sa677@njit.edu.