Using Alternative Date to Predict Key Performance Indicators

Abstract

The integration of robust data analysis into the financial markets has allowed firms to develop more efficient and accurate algorithms for predicting the behavior of individual firms. This analysis has been most effective with large and comprehensive data sets. However, the ease of integration for these types of data sets has led to their widespread adoption has diminished their usefulness in the financial markets. The most valuable data then exists in forms that are difficult to use and have not been used by the majority of the firms in the market. The purpose of this project is to integrate foot traffic data from individual stores into a quantitative and fundamental framework for predicting the values of significant KPIs for public firms.

Keywords: Clustering, Correlations, Foot Traffic, Key Predictive Indicators, Quantitative Analysis, Fundamental Analysis, Weather Data

JEL Classification: C80, G10, G12