Publication Date

1-2020

DOI

10.26716/redlands/master/2020.1

Committee Chair

Fang Ren, Ph.D.

Committee Members

Douglas M. Flewelling, Ph.D.

Abstract

Understanding the trends of home values through space and time can provide valuable economic and financial insights for policy makers, real estate professionals, and prospective home buyers. To better understand these trends, Johannes Moenius desired an effective and efficient method of visualizing and statistical analyzing historical home transaction data, as well as machine learning models to predict home values from assessor features. To achieve these requirements, a method of processing historical transaction point data into space-time cubes and corresponding hot spot analyses was developed, and multiple Python scripts were written to apply assessor data to Multiple Linear Regression Model and Random Forest Regression machine learning models. While the method developed for creating and analyzing space-time cubes proved effective, the machine learning models developed resulted in large errors in their home value predictions, likely due to the limited information of the assessor data used.

PolkBreck_Poster.pdf (2324 kB)
Poster

Creative Commons License

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.

Share

Article Location

 
COinS