EDA on King Country house sales

This Exploratory Data Analysis (EDA) was a three day task in my Data Science Bootcamp at neue fische.

Project time
2023-08-09 - 2023-08-11

Business case

My client Amy Williams is a house seller. She "works" as an Italian mafiosi and sells several central houses (top10%) over time. Now she is looking for average outskirt houses over time to hide from the FBI.

Business understanding

1. Find central top 10% houses, show prices over time
2. Find outskirt average houses, show prices over time

Goal

• Find recommendations for my client
• clean EDA-notebook
• presentation for "stakeholders" (bootcamp participants and coaches)

Dataset overview

The King County Housing dataset contains information about 22000 home sales in King County (USA) from 2014_05 until 2015_05. The description of the column names can be found in the column_names.md and feature_description.md file.

Recommendations and findings

• Top 10% houses in the center
  • sell in autumn
  • don’t sell in spring
  • median price ($): 2,500,000
• Average houses in the outskirts
  • buy in December or January
  • sell in April
  • median price ($): 419,980
• Average houses in the north
  • cheaper than average outskirts
  • median price ($): 405,000

Future Work

• EDA deep dive:
  • select fitting houses
  • refine customer needs
• clean notebook and repo
• optional:
  • write python scripts
  • unit testing

Technologies and methods used

• Jupyter Notebooks for EDA
• google slides for presentation

Requirements

• pyenv
• python==3.11.3

Set up environment

This repo contains a requirements.txt file with a list of all the packages and dependencies you will need. In order to install the environment you can use the following commands:

pyenv local 3.11.3
python -m venv .venv
source .venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt