README

1 Setup

1.1 Set up the repo for the Jupyter Notebook

Clone this repository.

Start by cloning the repo, to create a local copy of this folder in your machine.

In the terminal:

cd ~/Documents/

git clone https://github.com/audantic/take-home-project-data-analyst.git

cd take-home-project-data-analyst

1.2 Set up Docker

Install Docker: https://www.docker.com/community-edition#/download

This may take a while depending on download speeds.

docker pull jupyter/scipy-notebook

1.3 Start the notebook

docker run -it --rm -p 8888:8888 -v $(pwd):/home/jovyan/work jupyter/scipy-notebook

In the terminal, it will say something like:

$ docker run -it --rm -p 8888:8888 -v $(pwd):/home/jovyan/work jupyter/scipy-notebook
/usr/local/bin/start-notebook.sh: ignoring /usr/local/bin/start-notebook.d/*

Container must be run with group root to update passwd file
Executing the command: jupyter notebook
[I 20:38:18.195 NotebookApp] The Jupyter Notebook is running at:
[I 20:38:18.195 NotebookApp] http://[all ip addresses on your system]:8888/?token=c08dd7b6e98ac6658203af4426626e1ae2514cdb8c60c864
[I 20:38:18.195 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).
[C 20:38:18.196 NotebookApp] 
    
    Copy/paste this URL into your browser when you connect for the first time,
    to login with a token:
        http://localhost:8888/?token=c08dd7b6e98ac6658203af4426626e1ae2514cdb8c60c864

Copy/paste this URL into your browser.

Once the notebook is running, go into the work directory and open the "project.ipynb"

2 Dataset

The dataset is a portion of what we use for building predictive models.

Columns:

Overview Each row represents a home sale.

We try to predict both home sales, and types of sales.

When audantic_target=1, this is the type of transaction we are trying to predict.

Group	Column	Description
Response Variable	audantic_target	1 or 0, if we are trying to predict
ID	pid	property unique id
ID	did	sales document unique id
ID	sfr	bool for is single family home
Geographic	fips	county id
Geographic	zipcode	zipcode
Property	square_footage
Property	year_built
Property	equity
Property	estimated_value
Property	tax_assessed_value
Property	length_of_ownership
Demographic	est_household_income_val	estimated income
Demographic	mosaic_hh_val	demographic variable for home
Demographic	mosaic_zip4_val	demographic variable for neighborhood
Demographic	mosaic_diff	difference between home and neighborhood demographics
Sale Descriptors	sale_amount	home sale price
Sale Descriptors	seller_name
Sale Descriptors	buyer_name
Sale Descriptors	seller_occupied	if seller lived in the house
Sale Descriptors	document_type	Warranty Deed, Grant Deed, etc.
Sale Descriptors	distress_circumstance
Sale Descriptors	transfer_type

Objective

The objective of the take home project is to demonstrate how you approach data analysis.

• What is your thought process?
• What technical skills do you have?
• Can you generate insights from the given dataset?

You can create any or all of the following:

• Summary statistics
• Segment analysis
• Build charts
• Build histograms

Suggestions for starting points:

• Is there a difference between zipcodes?
• Is there a difference by year built buckets?
• Is there a difference in the property features between audantic_target=1 and =0?

Reference

Data visualization with Pandas