Creation of multiple charts and tables for representing useful insights/findings.
The charts
used should be in-line with the objective that you wish to convey to the Senior Management.
[Mandatory 8 types of charts/tables from any of the following: text table, bar
chart(multiple/stacked/side by side), bubble chart, treemap, Pareto chart, scatterplot,
Wordcloud, line plot, histogram, boxplot, circle views, heatmap, highlighted tables. No
restrictions on the upper limit of the number of charts/tables to be used] . **Please note the
charts should represent useful insights**
Objective: The project's goal is to conduct a comprehensive analysis of a car insurance dataset
using Tableau. The analysis and resulting recommendations will align with the business
objectives of a car insurance company. The primary objective involves delving into the dataset,
generating various plots, and extracting meaningful insights and findings. Ultimately, the aim is
to construct a compelling storyboard for presentation to Senior Management, proposing
decisive actions to enhance the current claims system.
Assumptions:
1) Car Owner and Driver is a same person
2) All amounts are in US Dollars
3) There are no specific fields to identify the unique customers. Hence each claim is assumed to
involve different customers.
Data Dictionary:
ID Identification Variable
KIDSDRIV Number of teenagers among the car owner's children who can drive a car.
BIRTH Date of birth of the driver
HOMEKIDS No of childern the car owner has
YOJ Years on Job. How many years has the owner of the car been working?
INCOME Income of the driver
PARENT1 Is the car owner a Single Parent
HOME_VAL Value of the house owned by the car owner
MSTATUS Marital status of the car owner
GENDER Gender of the driver
EDUCATION Maximum Education level of the driver
OCCUPATION Occupation of the driver
TRAVTIME Time taken to get to work on an average
CAR_USE Purpose of using the car
BLUEBOOK What is the worth of the car. Value of the Vehicle(in dollars)
CAR_TYPE Car type
OLDCLAIM Total claim (in past 5 years - in dollars)
CLM_FREQ Number of claims (in past 5 years)
CLM_AMT If car was in a crash, what is the currently claimed amount(in dollars)
CAR_AGE Age of car
URBANICITY Where the car is being driven primarily
Data Preparation
Make Education Dimension as an ordinal variable
The Education variable is a categorical variable. But actually it is ordinal variable. Hence the 5
values were manually sorted as follows
<High School
High School
Bachelors
Masters
PhD
Calculation of Age measure from birth date
Using the formula below, we calculated the age of each driver/owner
DATEDIFF('year', [Birth], TODAY())
Data Exploration
The database is of totally 7647 claims.
Average age of the driver is 68.75 years
Median age of the driver is 69 years
Average age of the car is 7.9 years
Average bluebook value is USD 15,176
Average income of the driver is USD 57,827
Average claim amount is USD 1,482
Average Income by gender and education: We can see there is a significant difference in the
income of the car owners by their education type. PhD holders earn highest while the
<Highschool car owners earn the lowest.
Average claim amount by car type
The average claim amount by Panel Truck owners is highest while the minivan claims avg
amount is the lowest
Age Vs Claim Amount chart: There is no significant correlation found between age of the car
owner and the claim amount.
The average income of doctors is highest whereas the average income of a student is lowest. It
is surprising to see that all occupation holders buy all type of cars
Older the car lower is the claim amount. This negative correlation is strongest in the minivan
car types and weakest in SUVs
Pareto Chart
The above pareto chart shows that the total claim amount of blue collar professional is the
highest and that of doctors is the lowest. This shows better paying and better status
professional people tend to be safe drivers.
Average claim amount of commercial vehicles is higher whereas that of the private vehicles is
lower
Number of kids driving does not have a significant effect on the average claim amount
The above tree map shows that the commercial used cars owned by blue collar people have
more average claim amount
Creating a calculated field. The calculated field should add some meaningful value and should
be in-line with your storyboard which you will create in this project. ** Please make sure that
you specify wherever calculated field has been used**
. The calculated field should add some meaningful value and should be in-line with your
storyboard which you will create in this project. ** Please make sure that you specify wherever
calculated field has been used**
Age of the car owner:
Using the formula below, we calculated the age of each driver/owner
DATEDIFF('year', [Birth], TODAY())
Age gives us a good insight on how the age of the owner affects the claim amount.
Age group calculation
Using the formula below, we calculated the age of each driver/owner
DATEDIFF('year', [Birth], TODAY())
Age gives us a good insight on how the age of the owner affects the claim amount.
Claim Frequency Calculation
Using the formula below, we calculated the Claim frequency
SUM([Clm Freq])
Tableau Project Link
https://public.tableau.com/app/profile/rajendra.ladda/viz/RajendraLaddaDVTprojectCarInsura
nce/Dashboard1?publish=yes
