Faculty Advisor

Humi, Mayer


This project investigates the possible short term prediction of stock prices using historical data of the stock and market in general. To this end, we created several models from real data and integrated multiple tools such as autocorrelation, least square linear fit, fourier series, moving averages, correlation with Nasdaq and Dow Jones. The models provided good predictions for over half of the stocks under consideration for ten business days. Further refinements of the model should be considered.


Worcester Polytechnic Institute

Date Accepted

February 2018

Project Type

Interactive Qualifying Project


Restricted-WPI community only

Advisor Department

Mathematical Sciences

Available for download on Saturday, February 27, 2021

Your accessibility may vary due to other restrictions.