Multivariable regression is used to analyze a dataset of properties compiled from the Multiple Listing Service (MLS) - a service used by realtors to list and find properties. The dataset includes information regarding each home’s features, such as number of bathrooms, bedrooms, square footage, acreage, exterior material, number of fireplaces, types of flooring, etc. My results demonstrate which of the observed features show statistical/practical significance in determining the final selling price of a home. The data is split into two equal halves and the same regression equation is run on both halves to internally replicate the study. The study is internally replicated to address the concern over p-hacking in the academic community and ensure the integrity of my findings. Previous research in this area shows that the inclusion of premium features - such as brick or stucco exteriors, decks, and hot tubs - positively influence a home’s selling price. This project contributes to this literature by establishing a curb appeal rubric and applying this rubric to the dataset. Curb appeal encompasses the attractiveness of a property when viewed from the street. The results of this project will demonstrate the potential impact of both a home’s curb appeal and features on its value. This topic could be of interest to many current/potential homeowners in the community and would be of particular interest to realtors/builders.
