Date of Award

5-2015

Degree Type

Thesis

Degree Name

Departmental Honors

Department

Economics and Finance

Abstract

This paper uses a variety of multiple regression analysis techniques to attempt to answer whether a direct relationship exists between Utah's employee wages and Utah's residential real estate values. Unexpected declines in real estate values can have seriously negative impacts on businesses, individuals, and local governments in Utah. Conversely, unexpected increases represent missed opportunities. Researchers have used various statistical and mathematical methods to explain or predict changes in real estate values, but no method has consistently predicted values for a long period of time or across multiple geographical areas. This paper focuses on exploring the relationship between variables in Utah and uses a linear probability model with nine explanatory variables to attempt to explain trends in quarterly data from the Utah Housing Price Index over the last sixteen years.

Initially, the regression returned promising numbers, but the results were misleading. Due to nonstationary data, high levels of autocorrelation, and other issues related to time-series data, the regression results were spurious, and no useable conclusions were drawn from the first model. In an attempt to correct for autocorrelation and the nonstationarity, the variables were transformed using a Prais-Winston transformation. Again, the results appeared promising. After multiple tests for stationarity and autocorrelation, however, the results were found to be autocorrelated and spurious.

That being said, the time spent reading complex papers, gathering reliable data, researching advanced regression methods, transforming variables, re-specifying models and analyzing results has been a great help and will contribute to a solid statistical foundation in the future. Research opportunities are available in the future when higher level statistical methods are learned. A relation between Utah's wages and Utah's real estate values may exist, but the statistical methods necessary to create the proper model are beyond the scope of this paper.

Download

Included in

Finance Commons

Share

COinS

Copyright for this work is retained by the student. If you have any questions regarding the inclusion of this work in the Digital Commons, please email us at .

 

Faculty Mentor

Ryan Bosworth

Departmental Honors Advisor

Frank Caliendo