California Housing Prices - Median house prices for California districts derived from the 1990 census.
Over the years, transactions under the real estate industry have increased many folds. The pandemic has further driven the middle income populations mindset towards acquiring personal residential property. Housing prices, which depend on the different parameters associated with the house, are an important metric for closing and finalizing any such residential real estate transaction.
In this project, we attempt to study the effect of different variables associated with a house on the price of the house. While the data being used for this study is old, we present this project as an academic study on the utility of statistics and regression to model and predict the housing prices.
We plan to implement the concept learned from the STAT 420 course at UIUC to do preliminary statistical analysis on the 1990 California House Prices and plan to narrow down on an accurate mode to predict the house prices based on certain relevant parameters.
The dataset which is being used in this study is publicly available and has been taken from Kaggle. The dataset is named as "California Housing Prices" and can be accessed as a csv file at https://www.kaggle.com/camnugent/california-housing-prices.
Originally, the dataset was made available at the Statistics Library in CMU on the StatLib platform as housing.zip. The original dataset can be access at http://lib.stat.cmu.edu/datasets/ under the names housing.
The contains information from the 1990 California census and pertains to the houses found in a given California district. It contains some of the summary statistics about the houses based on the 1990 California census. As stated above, although the dataset is old, we wish to present this project as an academic study on the utility of statistics and regression to model and predict housing prices.
This data was initially featured in the following paper: Pace, R. Kelley, and Ronald Barry. "Sparse spatial autoregressions." Statistics & Probability Letters 33.3 (1997): 291-297.
The California Housing Prices dataset provides the median house prices for California districts derived from the 1990 census data. It contains one row per census block group. A block group is the smallest geographical unit for which the U.S. Census Bureau publishes sample data (a block group typically has a population of 600 to 3,000 people).
It contains a total of 10 columns which include the median house prices and other variables which might be associated with the price.
The data contains almost 2060 rows. We will build a model to predict is "Median-House-Value" and the rest are columns will be used as predictor variables. The dataset is not cleaned so we will be pre-processing it before using it for modeling.
**The following are data points from the dataset: **
-
1.longitude: A measure of how far west a house is. The higher value of this parameter means farther west. The California longitude value ranges from: 114° 8' W to 124°.
-
2.latitude: A measure of how far north a house is. The higher value of this parameter means farther north. The California Latitude value ranges from: 32° 30' N to 42° N
-
3.housingMedianAge: Median age of a house within a block (a block has population of around 600 to 3000 people). The lower number for this parameter means the building is newer.
-
4. totalRooms: Total number of rooms among all houses within a block.
-
5. totalBedrooms: Total number of bedrooms among all houses within a block.
-
6. population: Total number of people residing within a block.
-
7. households: Total number of households, a group of people residing within a home unit, for a block.
-
8. medianIncome: Median income for households within a block of houses (measured in tens of thousands of US Dollars)
-
9. medianHouseValue: Median house value for households within a block (measured in US Dollars)
-
10. oceanProximity: Location of the house w.r.t ocean/sea.
Ocean_proximityindicating (very roughly) whether each block group is near the ocean, near the Bay area, inland or on an island. This parameter will allows us include and interpret the categorical variable while regressing the dataset.
We choose this model for our analysis because we want -
-
Hands on expedience with real life dataset. We chose this dataset because it is a medium sized yet not small or too large. This dataset satisfy all the project requirements.
-
The motivation of choosing this dataset is to practice all techniques learned in STAT420.
-
Discover how applied statistics can help us building the most accurate model and predicting the subject of our study(House Values).
- Advantages of an accurate model
- When Closing a sale it is important to quote a good price