Linear Regression modelling on housing data
In this project, I shall be analysing housing data in King County, Washington to investigate house pricing determinants.
As a potential investor interested in venturing in the real estate business and would like to mitigate risks that might be encountered in the industry and plan ahead for what is likely to go wrong or what to do right. This exploration of the housing data in the city is intended to give a feel of what the real estate business in the City might look like, what factors influence sales and pricing.
This project uses the King County House Sales dataset. The Data has several variables that can be used as predictors or seem to have influence on the sale price of houses. The
I used Python and Jupyter Notebooks for data analysis, cleaning and inferences. I started by loading the data, exploring the contents and did some data cleaning to remove. There is also a Tableu viz in the folders just for the visual analysis.
The data has variables that describe location, quality and quantity of the houses. Given the county is in a coastal area, things like waterfront properties and views are also factored in as they might influence price in a way.
I used linear regression to investigate the relationship in variables as it is the most comprehensive method.
- If all other factors were 0, price would be 7071780.6
- Luxury and mansion categories affect the price more
- Houses with very good conditions are more likely to increase price
- Having a Waterfront property is statisticaly more significant than not having at alfa = 0.05
- the number of floors and bathrooms has a great effect in the price
Waterfront properties are generally hotcakes. The bigger the living space, the higher the price. If you have good quality houses with top of the line finishes, the money will also be top of the table. Renovations are always needed on old property in order for the to compete at market levels.
I found that location, views and quality and measurement/ size of the house have a huge influence on the price. With size of the house and area taking the largest slice of the pie. If it is coastal, invest in waterfront. The views are really where its at.