PGPDS(Post Graduation Program
in Data Science- Mar. 25
Anurag Kumar Singh
Contents:-
1. List of Figures.
2. Understanding the data and its Structure.
3. Data Checks and treating irregularities.
--Checking missing data fields.
--Checking data consistency
--Checking Statistical data.
4. Univariate Analysis.
5. Bivariate Analysis.
6. Key Questions and recommendations.
1. List of Figures.
Fig 1.1 – Duplicate.
Fig 1.2 – Missing Fields.
Fig 1.3 – Data Consistency.
Fig 1.4 – Statistical Summary.
Fig 1.5 – Bar graph Gender
Fig 1.6 – Bar graph Profession.
Fig 1.7 – Bar graph Marital_status.
Fig 1.8 – Bar graph Education.
Fig 1.9 – Bar graph Personal_loan.
Fig 1.10 – Bar graph House_loan.
Fig 1.11 – Bar graph Partner_working.
Fig 1.12 – Bar graph Vehicle Type.
Fig 1.13 – Histogram-Age.
Fig 1.14 – Histogram-Salary.
Fig 1.15 – Histogram-Partner Salary.
Fig 1.16 – Histogram-Total Salary.
Fig 1.17 – Histogram-Price.
Fig 1.18 – Pairplot.
Fig 1.19 – Correlation Table.
Fig 1.20 – Heatmap.
Fig 1.21 – Car Preference for male and female.
Fig 1.22 – Car Preference as per Profession.
Fig 1.23 – Car Preference as per Profession and Gender.
Fig 1.24 - Car Total amount spent on vehicle by Gender.
Fig 1.25 - Total amount spent on vehicle by Personal Loan.
Fig 1.26 - Influence of Partner working on purchase of High-priced cars.
2. Understanding the data and its Structure.
Austo Motor Company is a leading car manufacturer specializing in SUV, Sedan, and Hatchback models. In its recent board meeting,
concerns were raised by the members on the efficiency of the marketing campaign currently being used. The board decides to rope in
an analytics professional to improve the existing campaign.
• Data Contains 1581 rows and 14 coulmns.
• Data Description:-
Column Name Data Type
Age The age of the individual in years.
Gender The gender of the individual, categorized as male or female.
Profession The occupation or profession of the individual.
Marital_status The marital status of the individual, such as married &, single
Education The educational qualification of the individual Graduate and Post Graduate
No_of_Dependents
The number of dependents (e.g., children, elderly parents) that the individual supports financially.
Personal_loan A binary variable indicating whether the individual has taken a personal loan "Yes" or "No"
House_loan A binary variable indicating whether the individual has taken a housing loan "Yes" or "No"
Partner_working A binary variable indicating whether the individual's partner is employed "Yes" or "No"
Salary The individual's salary or income.
Partner_salary The salary or income of the individual's partner, if applicable.
Total_salary The total combined salary of the individual and their partner (if applicable).
Price The price of a product or service.
Make The type of automobile
3. Data Checks and treating irregularities.
• There are no duplicates in the data.
Checking missing fields.
Fig.1.1 Duplicate
• We check the missing values and found that data type Gender and
Partner_salary has missing values.
Fig.1.2 Missing Fields
• We found that there are 53 null values in Gender. We can drop the Nan values as it affect other variable. Creating new
columns or category is not suitable as its only 3% of data.
So we replaced the Nan value with mode.
• Partner salary contains 106 missing values. This variable is dependent upon other two variable which are Marital
status and Partner_Working is “Yes” we ensured that all the records must be mention “Married” under Marital Status.
• Rows where partner_salary is null is obtained from subtracting salary from total,i.e., Partner_salary= Total Salary-
Salary. Also whenever Partner working is “no” the value of Partner_working column would be 0.
Checking Data Consistency.
• We check unique values for categorical data and found that gender column is inconsistent. It contains
spelling error that mentions:- ‘Male’, ‘Femal’, ‘Female’, ‘Femle’.
• ‘Femal’ and ‘Femle’ where replaced with ‘Female’.
Fig.1.3 Data Consistency
Checking Statistical data.
Fig.1.4 Statistical Summary
Observation:-
• Mean age is 31.9 years ands max is 54 years.
• Average number od dependent is 2.45 ~ 2-3dependents .
• Average Salary is 60392.22.
• Average Partner_salary is 19233.77.
• Average Total salary is 79625 with max 171000.
• Average price at which vehicle bought is 35597.72 and max at 70000.
4. Univariate Analysis.
a) There are more males (n=1252) in data than females (n=329).
Fig.1.5 Gender
b) We found there are more salaried people in data set.
Fig.1.6 Profession
c) We analysed that most of the people are married.
Fig.1.7 Marital_status
d) In the data more number of people are postgraduate.
Fig.1.8 Education
e) We could see 50% of people have personal loan.
Fig.1.9 Personal Loan
f) 33% of people have house loan.
Fig.1.10 House Loan
g) We observed that more that 800 people have working partner.
Fig.1.11 Partner Working
h) Most of the people have Sedan followed by Hatchback.
Fig.1.12 Vehicle Type
i) We observed that 60% of people have age less than 30.
Fig.1.13 Histogram-Age
j) We found salary has normal distribution curve.
Fig.1.14 Histogram-Salary
k) Partner Salary is right skewed. Most of the values are piled up on the left (near 0 to 10,000) and the tail (a few values)
stretches far to the right (up to 70,000–80,000).
Fig.1.15 Histogram-Partner Salary
l) Total Salary is right skewed . The curve (KDE line) leans left and trails right
Fig.1.16 Histogram-Total Salary
m) Price is right skewed since the curve (KDE line) rises steeply and then slowly tapers off to the right.
Fig.1.17 Histogram-Price
5. Bivariate Analysis.
•Conducted pairplots and calculated correlation for continuous
variables.
•Identified a linear relationship between:
• Age and Salary
• Age and Total Salary
• Age and Price
• Salary and Total Salary
• Price and Salary
• Partner Salary and Total Salary
• Total Salary and Price
Fig.1.18 Pairplot
Correlation Table
Fig.1.19 Correlation Table
Correlation Matrix Heatmap.
Below are correlation observed:-
• Age and price have strong positive
correlation.
• Salary and total salary has positive
correlation.
• No of dependant and price has
very weak negative correlation.
•
Fig.1.20 Heatmap
6. Key Questions.
Observation:-
• Mens buy more car in comparison to women.
• Men prefers to buy Sedan and hatchback
than SUV.
• Women pefer SUV more that Sedan and
Hatchback.
Recommendations:-
Market strategy should be followed on more
focused product depending on
group preference. Discount and promotional
offer should be rolled out for more sales.
Fig.1.21 Car Preference for male and female.
Observation:-
• In Salaried person people prefer Sedan and
SUV as compared to business.
• For salaried most preferred is Sedan followed
by Hatchback.
Recommendation:-
• Emphasize Sedan features like comfort, fuel
efficiency, and technology in ads.
• Highlight loan offers (EMI plans) and fuel
savings — since they seem to prefer
Sedans and SUVs.
Fig.1.22 Car Preference as per Profession.
Observations:-
The statement made by Sheldon Cooper is False:-
• Salaried male prefer Sedan over SUVs.
• Salaried female prefers SUVs the most followed by
Sedan.
Fig.1.23 Car Preference as per Profession and Gender.
Observation:-
• We observed that female spend more on buying
cars than the male.
Recommendations:-
• Marketing team should focus more in females
and work to boost sales and profit.
Fig.1.24 Car Total amount spent on vehicle by Gender
Observations:-
• We found that people do not having personal
loan spend more in buying car on average than
people with personal loan.
Recommendations:-
• Focus marketing and sales efforts on people who
do not already have a personal loan.
• Even though people with personal loans spend
less, you can encourage them to buy by offering
special car loan deals.
Fig.1.25 Total amount spent on vehicle by Personal Loan
Observations:-
• Having a working partner does not
lead to purchase of high price cars.
Recommendation:-
• Marketing efforts do not need to be
segmented based on whether people have
a working partner or not.
Fig.1.26 Influence of Partner working on purchase of High-priced cars
7. Recommendations.
• Focus marketing on customers with strong financial health (e.g., no personal loans, good credit scores).
• Focus of males to increase overall sales figures and for Sedan and hatchback on females and offer discount.
• Provide low-interest loans, extended EMIs, zero down payment schemes.
• Offer limited-time bonuses: free accessories, free insurance, free servicing.
• Create a strong website where customers can explore models, customize cars, and even apply for financing online.
• Use customer data (age, income, profession) to create personalized offers.
