General DG R Assessment Task 2018 Instructions

User Manual:

Open the PDF directly: View PDF PDF.
Page Count: 1

DownloadGeneral DG R Assessment Task 2018 - Instructions
Open PDF In BrowserView PDF
Domestic & General Group Ltd.
Data Science & Customer Analytics Team
Author: Jeffrey Fung
Date: 01/10/2018

Assessment Task
Please use R (or another language of your choice) to answer the below questions concisely using
statements, graphs and tables where applicable, making use of any packages of your choice. R
script(s), R Markdown or Jupyter Notebook are all acceptable. You have 48 hours to complete this
test so please do not feel rushed and it is “open book”, so feel free to use any online or offline
resources during this time.
After completion please email back an archived folder with all your answers (data, scripts, graphs
etc.) and your initials in the file name.
If anything is unclear or you are unsure how to answer a question, please note this down in the
email and do not worry.
Data:
Two files are provided, Offers_sent.csv and Offers_accepted.csv. They respectively contain
data on offers sent to, and accepted by, customers for a specific appliance.
Offers_sent.csv variables are as follows:
A. OfferContactDate – Contact date, DD/MM/YYYY;
B. CustomerID – Anonymized unique customer ID;
C. ApplianceID – Anonymized unique appliance ID;
D. Brand – Anonymized brand code;
E. V1-V8 – Anonymized categories and counts.
Offers_accepted.csv variables are as follows:
A. OfferAcceptanceDate – Offer acceptance date, DD/MM/YYYY;
B. CustomerID – Anonymized unique customer ID;
C. ApplianceID – Anonymized unique appliance ID;
Tasks:
a) The conversion rate is defined as the number of offers accepted divided by the number of
offers sent. Calculate the conversion rate for each brand and assess whether there is a
statistically significant difference between them.
b) Graphically present how the total acceptance of offers develops over time, i.e. how many
offers have been accepted after a specified time from being sent (daily or weekly graph).
c) Using any technique, build a simple predictive model to predict whether an offer will be
accepted using columns V1-V8. Briefly explain your choice of model, and decide which
variables are most important in the model. Furthermore, examine the model and report
on an accuracy metric(s) of your choice. Please do not spend too much time optimising
model parameters.
d) V7 is the treatment category and currently, the business only applies treatment A. The
business has decided to introduce new treatment categorical values to V7, factors B, C, and
D. Design an experiment to assess the potential impact of the new factors on conversion
rates and include which statistical methods you would use. Please state any assumptions
made. (R is not required here, you can answer this question in comments)
Company Confidential, Copyright © 2018 Domestic & General Group Ltd. All rights reserved.
TVX 1.0

Page 1 of 1
Source Exif Data: 
File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.7
Linearized                      : No
Page Count                      : 1
Language                        : en-CA
Tagged PDF                      : Yes
XMP Toolkit                     : 3.1-701
Producer                        : Microsoft® Word 2016
Title                           : General Document
Creator                         : Jeffrey Fung
Creator Tool                    : Microsoft® Word 2016
Create Date                     : 2018:10:02 12:40:25+01:00
Modify Date                     : 2018:10:02 12:40:25+01:00
Document ID                     : uuid:B1EEDD0E-D7B1-46BB-A9D1-DF3A24C0F751
Instance ID                     : uuid:B1EEDD0E-D7B1-46BB-A9D1-DF3A24C0F751
Author                          : Jeffrey Fung
EXIF Metadata provided by EXIF.tools

Navigation menu