Project: Predict future sales of Walmart
Table of Contents
1. Introduction 1
2. Project assignments 2
Assignment 1: Frame the problem 3
Assignment 2: Collect raw data 3
Assignment 3: Process the data 3
Assignment 4: Explore the data 3
Assignment 5: Perform in-depth analysis 4
Assignment 6: Communicate results 4
3. Planning 4
4. Coaches 4
1. Introduction
In this project you’ll participate in a data science competition at Kaggle.com:
https://www.kaggle.com/c/walmart-recruiting-store-sales-forecasting/
This competition took place in 2014 and was intended to find suitable candidates for data science jobs at Walmart. The competition is still available even though they might not be recruiting anymore.
You’ll find the following text when you click the link to the competition:
“One challenge of modeling retail data is the need to make decisions based on limited history. If Christmas comes but once a year, so does the chance to see how strategic decisions impacted the bottom line.
In this recruiting competition, job-seekers are provided with historical sales data for 45 Walmart stores located in different regions. Each store contains many departments, and participants must project the sales for each department in each store. To add to the challenge, selected holiday markdown events are included in the dataset. These markdowns are known to affect sales, but it is challenging to predict which departments are affected and the extent of the impact.
Want to work in a great environment with some of the world's largest data sets? This is a chance to display your modeling mettle to the Walmart hiring teams.
This competition counts towards rankings & achievements. If you wish to be considered for an interview at Walmart, check the box "Allow host to contact me" when you make your first entry.
You must compete as an individual in recruiting competitions. You may only use the provided data to make your predictions.
You are provided with historical sales data for 45 Walmart stores located in different regions. Each store contains a number of departments, and you are tasked with predicting the department-wide sales for each store.
In addition, Walmart runs several promotional markdown events throughout the year. These markdowns precede prominent holidays, the four largest of which are the Super Bowl, Labor Day, Thanksgiving, and Christmas. The weeks including these holidays are weighted five times higher in the evaluation than non-holiday weeks. Part of the challenge presented by this competition is modeling the effects of markdowns on these holiday weeks in the absence of complete/ideal historical data.
stores.csv
This file contains anonymized information about the 45 stores, indicating the type and size of store.
train.csv
This is the historical training data, which covers to 2010-02-05 to 2012-11-01. Within this file you will find the following fields:
Store - the store number
Dept - the department number
Date - the week
Weekly_Sales - sales for the given department in the given store
IsHoliday - whether the week is a special holiday week
test.csv
This file is identical to train.csv, except we have withheld the weekly sales. You must predict the sales for each triplet of store, department, and date in this file.
features.csv
This file contains additional data related to the store, department, and regional activity for the given dates. It contains the following fields:
Store - the store number
Date - the week
Temperature - average temperature in the region
Fuel_Price - cost of fuel in the region
MarkDown1-5 - anonymized data related to promotional markdowns that Walmart is running. MarkDown data is only available after Nov 2011, and is not available for all stores all the time. Any missing value is marked with an NA.
CPI - the consumer price index
Unemployment - the unemployment rate
IsHoliday - whether the week is a special holiday week
For convenience, the four holidays fall within the following weeks in the dataset (not all holidays are in the data):
Super Bowl: 12-Feb-10, 11-Feb-11, 10-Feb-12, 8-Feb-13
Labor Day: 10-Sep-10, 9-Sep-11, 7-Sep-12, 6-Sep-13
Thanksgiving: 26-Nov-10, 25-Nov-11, 23-Nov-12, 29-Nov-13
Christmas: 31-Dec-10, 30-Dec-11, 28-Dec-12, 27-Dec-13
2. Project assignments
You’ll go through all phases of a data science project:
https://ajgoldstein.com/2017/11/12/deconstructing-data-science/
Assignment 1: Frame the problem
Carefully read the competition description on Kaggle.com and execute the following tasks:
1.Define the problem that Walmart wants you to solve.
2.Describe the decisions that Walmart can take after the delivery of your work
Assignment 2: Collect raw data
1.Create one profile for your team on Kaggle.com. Let us know what the username of your team is:
User name:………………
2.Download the data set and upload the data set to the files folder in your MS Teams channel
Assignment 3: Process the data
1.Import the data set into a Jupyter Notebook as dataframes
2.Examine the data at a high level:
a.Understand every column
b.Identify errors, missing values & corrupt records (make sure to check this even though it’s quite a clean data set)
3.Clean the data (make sure to check this even though it’s quite a clean data set)
a.Throw away, replace, filter corrupt/ error prone / missing values
Assignment 4: Explore the data
1.Merge the dataframes in such a way you can use it for exploratory data analysis
2.Play around with the data:
a.Plot the relations between each of the input variables and the output variable (sales).
b.Use statistics to identify significant variables and create relevant features.
Do your analysis for the total and at least 3 stores and 3 departments
Assignment 5: Perform in-depth analysis
1.Create a predictive model
a.Use the features that you found in assignment 4 and day, department and store
b.Use 3 types of model
2.Evaluate and refine the model
a.You may have to revert back to steps 2 and 3
b.Choose the model you want to use
c.Improve the model using cross validation and/or a validation set.
Assignment 6: Communicate results
1.Use power BI to show the results of your analysis. It should be understandable to business people with no data science background
a.Explain how you got to your results
b.Explain the results and how they should be interpreted considering the fact that your audience are not data scientists.
2.Give advice on the decisions to make by Walmart (as described in assignment 1)
a.Markdown strategy: How do you recommend to apply markdowns? Give your recommendations for 3 departments of 3 stores.
b.Holiday markdown strategy: How do you recommend to apply markdowns during the holidays? Give your recommendations for 3 departments of 3 stores.
c.Which indicators (CPI, Fuel Price, etc.) seem to be relevant for which departments of your three stores? How do you recommend Walmart should deal with changes on these indicators?
3. Planning
Date Time Topic Preparation
30th September 13:00 – 14:00 Introduction of project
7th / 8th October 11:00 – 16:00 Half hour of project coaching with group’ coach Submit draft version of your assignments for feedback
14th / 15th October 11:00 – 16:00 Half hour of project coaching with group’ coach Submit draft version of your assignments for feedback
28th / 29th October 11:00 – 16:00 Half hour of project coaching with group’ coach Submit draft version of your assignments for feedback
8th November 23:59 Deadline to submit project report and dashboard
11th November 9:30 – 15:30 Project presentations
4. Coaches
Group Coach Coaching half hour
1 Thomas Becker Wednesday 11:00 – 11:30
2 Thomas Becker Wednesday 11:30 – 12:00
3 Thomas Becker Wednesday 12:00 – 12:30
4 Erik van den Ham Wednesday 13:00 – 13:30
5 Erik van den Ham Wednesday 13:30 – 14:00
6 Erik van den Ham Wednesday 14:00 – 14:30
7 Erik van den Ham Wednesday 14:30 – 15:00
8 Raymond Hoogendoorn To be discussed with coach
9 Raymond Hoogendoorn To be discussed with coach
10 Raymond Hoogendoorn To be discussed with coach
12 Raymond Hoogendoorn To be discussed with coach