Sangram Keshari Patro

Business Report - 1 PG Program in Data Science and Business Analytics submitted by Sangram Keshari Patro BATCH:PGPDSBA.O.AUG24.B Contents 1 Context 2 2 Data Description 2 3 Data Overview 2 3.1 Importing necessary libraries and the dataset . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3.2 Structure and type of data 3 3.3 Statistical summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.4 Missing value treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.4.1 Treating 'Gender' column . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3.4.2 Treating 'Partner_salary' column . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Exploratory Data Analysis 4.1 4.2 4.3 Univariate Analysis 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.1 Numerical columns 4.1.2 Categorical columns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bivariate Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 8 13 4.2.1 Numerical variables 4.2.2 Categorical vs numerical variables . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 4.2.3 Categorical vs categorical variables . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 13 Key questions to ponder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 4.3.1 Do men tend to prefer SUVs more compared to women? . . . . . . . . . . . . . . . 25 4.3.2 What is the likelihood of a salaried person buying a Sedan? . . . . . . . . . . . . . . 25 4.3.3 What evidence or data supports Sheldon Cooper's claim that a salaried male is an . . . . . . . . . . . . . . . . . . . . 25 4.3.4 How does the the amount spent on purchasing automobiles vary by gender? . . . . . 26 4.3.5 How much money was spent on purchasing automobiles by individuals who took a personal loan? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 4.3.6 How does having a working partner inuence the purchase of higher-priced cars? . . 28 easier target for a SUV sale over a Sedan sale? 4.4 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 4.4.1 Some extra intersting questions to ponder How does the level of Education impact price of cars purchased? . . . . . . . . . . . 29 4.4.2 How does the total salary of customers inuence the price range of dierent car models (SUV, Sedan, Hatchback)? . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 32 1 Context They want to analyze the data to get a fair idea about the demand of customers which will help them in enhancing their customer experience. Suppose you are a Data Scientist at the company and the Data Science team has shared some of the key questions that need to be answered. Perform the data analysis to nd answers to these questions that will help the company to improve the business. 2 Data Description ˆ Age: The age of the individual in years. ˆ Gender: The gender of the individual, categorized as male or female. ˆ Profession: The occupation or profession of the individual. ˆ Marital_status: The marital status of the individual, such as married &, single ˆ Education: The educational qualication of the individual Graduate and Post Graduate ˆ No_of_Dependents: The number of dependents (e.g., children, elderly parents) that the individual supports nancially. ˆ Personal_loan: A binary variable indicating whether the individual has taken a personal loan "Yes" or "No" ˆ House_loan: A binary variable indicating whether the individual has taken a housing loan "Yes" or "No" ˆ Partner_working: A binary variable indicating whether the individual's partner is employed "Yes" or "No" ˆ Salary: The individual's salary or income. ˆ Partner_salary: The salary or income of the individual's partner, if applicable. ˆ Total_salary: The total combined salary of the individual and their partner (if applicable). ˆ Price: The price of a product or service. ˆ Make: The type of automobile 3 3.1 Data Overview Importing necessary libraries and the dataset The dataframe is printed. It has 1581 rows & 14 columns. 2 Figure 1: Dataframe 3.2 Structure and type of data Data is explored further. Data doesn't have any duplicate rows. Figure 2: Table depicting the datatype and Non-Null values in each column. This also denotes the columns having 'NaN' values. 3 3.3 Statistical summary Figure 3: Statistical summary of the data From this table we can observe that there are outliers in the 'Total_salary' column 3.4 Missing value treatment 3.4.1 Treating 'Gender' column Typos in the 'Gender' column were corrected. For instance, entries like 'Femal' and 'Female' were standardized to 'Female', while missing or NaN values were replaced with 'Unknown.' 3.4.2 Treating 'Partner_salary' column The boxplot and histogram for the 'Partner_salary' column indicate no extreme outliers. However, despite the absence of outliers, it's preferable to use the median to replace NaN values rather than the mean because the data distribution is not Gaussian (i.e., it may be skewed or non-symmetric). The median is a more robust measure of central tendency in such cases, as it is less inuenced by any skewness or irregularities in the data distribution. 4 Exploratory Data Analysis 4.1 Univariate Analysis 4.1.1 Numerical columns 4 ˆ Age Figure 4: Histogram and boxplot of 'Age' column 1. The histogram and kernel density estimate depict the age distribution of customers. Notably, this distribution is right-skewed, indicating that there are more younger customers and relatively fewer older ones. 2. Most customers appear to be in their younger years (between 20 and 30). This insight suggests that targeting marketing eorts toward this younger demographic could yield better results. 3. The box plot reveals the spread of ages within the sample population. The interquartile range (IQR) spans from approximately 25 to 38, covering the middle 50% of ages. The median age is 29. Targeting this age group might yield the highest engagement, as this age group constitutes the bulk of the customer base. ˆ No_of_Dependents Figure 5: Histogram and boxplot of 'No_of_Dependents' column 1. The histogram displays the count of individuals based on the number of dependents they have. The tallest bars correspond to individuals with 2 or 3 dependents. Austo Motor Company can leverage this information to tailor their marketing strategies eectively related to family-oriented campaigns or nancial planning. 2. In the box plot, we observe an outlieran individual with no dependents. The interquartile range (IQR), which represents the middle 50% of the data, spans from approximately 2 to 3 dependents.We also can observe that the median line is coinciding with the Q1. dependents is 4. 5 The highest reported number of ˆ Salary Figure 6: Histogram and boxplot of Salary column 1. The histogram shows the distribution of salaries. Most employees fall within the salary range around 60,000. The shape of the histogram is somewhat bell-shaped but skewed to the right, indicating fewer individuals with higher salaries and more with salaries below the peak. 2. If Austo's vehicles are considered luxury items, targeting higher salary brackets might be more eective. If aordability is key, focusing on the middle-income segment could improve campaign eciency. 3. The interquartile range (IQR) spans from 60,500 Rs. to 95,900 Rs. Allocate resources eectively based on customer segments. Tailor messaging to resonate with dierent income groups. ˆ Partner_salary Figure 7: Histogram and boxplot of 'Partner_salary' column 1. The histogram shows that most partners have salaries in the lower range (around 0 to 10,000 Rs). Fewer partners fall into higher salary brackets. Customized marketing messages can be crafted for each segment. Low-income partners might respond better to cost-saving benets, while high-income partners may prioritize performance or prestige. 2. The interquartile range (IQR) for the data lies between 0 and 38,300 Rs, with a median value of 25,600 Rs. 6 ˆ Total_salary Figure 8: Histogram and boxplot of 'Total_salary' column 1. The histogram shows the distribution of total salaries among customers. There are two prominent peaks in salary frequency: around 60,000 Rs. and 80,000 Rs. 2. The median salary 78,000 Rs. provides insight into the typical salary level.The interquartile range (IQR) spans from 51,900 Rs. to 71,800 Rs. We can observe the outliers in this dataset having high total salary. The mean total salary is 79,626 Rs. which is close to the median i.e. the data is not skewed. Treating outliers in Total_salary column All the values smaller than lower_whisker will be assigned the value of lower_whisker & all the values greater than upper_whisker will be assigned the value of upper_whisker. The plots after treatment is attached below. Figure 9: Histogram and boxplot of 'Total_salary' column after treatment of outliers Although the overall shape of the curve remains the same, the boxplot no longer includes outliers. 7 ˆ Price Figure 10: Histogram and boxplot of 'Price' column 1. The histogram shows that the majority of customers prefer cars priced between approximately 20,000 and 30,000 units. This price range has the highest concentration of data points. We can consider focusing marketing eorts on this segment, as it represents the largest customer base. 2. Most prices cluster around the median, suggesting a common price preference among customers. 4.1.2 Categorical columns ˆ Gender Figure 11: Barchart of 'Gender' column Observations Observations: The female customers. provided bar chart indicates a higher demand among male customers compared to Insights Insights: To enhance customer experience and improve marketing eciency, Austo Motor Company should consider tailoring campaigns to attract more female interest. 8 customers or further capitalize on the high male ˆ Profession Figure 12: Barchart of 'Profession' column Observations Observations: ˆ Notably, the number of Salaried individuals is signicantly higher, indicating a larger representation in the dataset. Insights Insights: ˆ Given the higher proportion of Salaried professionals, it appears that this group could be a promising target market for marketing campaigns. ˆ To improve campaign eciency, consider directing marketing eorts toward ˆ Marital_status Figure 13: Barchart of 'Marital_status' column 9 Salaried customers. Observations 1. The majority of customers are Married, with a count exceeding 1400. Insights 1. Marketing Strategy: Singles. Focus Sedan/Hatchback ads on Married individuals; tailor Hatchback campaigns for ˆ Education Figure 14: Barchart of ' Education' column Observations 1. The bar chart analysis indicates higher counts for Post Graduates compared to Graduates. Insights 1. This insight can guide Austo Motor Company in enhancing marketing strategies towards a more educated customer base, potentially improving customer engagement and sales. 10 ˆ Personal_loan Figure 15: Barchart of 'Personal_loan' column Observations The bar chart shows a nearly equal distribution of customers who have taken personal loans (Yes) and those who haven't (No). Insights Given the balanced distribution, the marketing campaign should target both loan and non-loan customers. Further analysis could explore campaign eectiveness across dierent customer segments (e.g., SUV, Sedan, Hatchback). ˆ House_loan Figure 16: Barchart of 'House_loan' column Observations The majority of customers do not have house loans. 11 Insights This information can guide targeted marketing eorts based on customer loan status. ˆ Partner_working Figure 17: Barchart of 'Partner_working' column Observations Customers with working partners have a higher count. Insights Consider partner employment status in marketing strategies.. ˆ Make Figure 18: Barchart of 'Make' column 12 Observations SUV, Sedan, and HatchSedan model exhibits the highest demand with around 700 units, followed by Hatchback approximately 580 units, and SUV at around 300 units. This data is crucial for Austo Motor The provided image shows a bar chart with counts for three car models: back. at The Company's marketing campaign eciency analysis. Insights Consider partner employment status in marketing strategies.. 4.2 Bivariate Analysis 4.2.1 Numerical variables ˆ Heatmap Figure 19: Heatmap of all numerical variables Key Insights: ˆ Age has a strong positive correlation with Price (0.8), indicating older customers prefer expensive cars. ˆ Salary and Price have a moderate positive correlation (0.41), showing higher salaries lead to more expensive purchases. ˆ Total salary shows a moderate correlation with Price (0.36), highlighting household income's inu- ence on car prices. ˆ No. of Dependents has a weak negative correlation with in expensive car purchases with more dependents. ˆ Pairplot 13 Price (-0.14), indicating a slight decrease ˆ Pairplot Figure 20: Pairplot of all numerical variables Observations Diagonal (Density Plots): ˆ Age: Skewed left, suggesting most customers are younger (around 25-35 years old). ˆ Salary: Mostly centered around 40,000-60,000, indicating a middle-income segment. ˆ Partner_salary: ˆ Total_salary: ˆ Price: Bimodal distribution with peaks at 0 (no partner salary) and around 25,000. Peaks around 50,000-100,000, indicating most households earn in this range. Slightly skewed to the right, with many cars priced between 20,000 and 50,000. Scatter Plots: ˆ Age vs Salary: Positive correlation. Younger customers tend to have lower salaries, while older customers have higher salaries. 14 ˆ Age vs Partner_salary: No clear pattern, suggesting partner salary does not vary signicantly with age. ˆ Age vs Total_salary: Positive correlation. Older customers generally have higher combined house- hold incomes. ˆ Age vs Price: Strong positive correlation. Older customers tend to buy more expensive cars. ˆ Salary vs Partner_salary: Weak relationship. Many customers with similar salaries have a wide range of partner salaries, including zero. ˆ Salary vs Total_salary: Strong linear correlation. Higher individual salaries lead to higher combined household incomes. ˆ Salary vs Price: Moderate positive correlation. Higher salaries lead to purchases of higher-priced cars. ˆ Partner_salary vs Total_salary: Moderate correlation, but many customers report a partner salary of zero, reducing the strength of this relationship. ˆ Partner_salary vs Price: Weak positive correlation, but no strong trend indicating a direct inuence of partner salary on car price. ˆ Total_salary vs Price: Positive correlation. Higher household income is associated with higher- priced car purchases. Key Insights ˆ Age strongly inuences Price and Total_salary, showing that older customers tend to buy more expensive cars and have higher combined incomes. ˆ Salary is a key factor for both Total_salary and Price, suggesting that customers with higher individual incomes are likely to buy higher-end models. ˆ Partner_salary has a limited direct impact on important in decision-making. 15 Price, indicating individual income might be more 4.2.2 Categorical vs numerical variables Figure 21: Violinplot of categorical vs numerical variables 16 Figure 22: Violinplot of categorical vs numerical variables 17 Observations Diagonal Plots (Violin Plots of Numerical Variables by Categorical Variables) Gender: ˆ Age: ˆ No_of_Dependents: ˆ Females slightly older on average. Salary: Females have fewer dependents, mostly 1-2. Males have more varied salaries but median salary is higher for female as older females are high in number in our dataset. ˆ Partner_salary: ˆ Total_salary: ˆ Price: Similar distributions across genders, many with no partner salary. Females have higher median total salaries, broader distribution. Females buy more expensive cars. Profession: ˆ Age: A larger proportion of business professionals fall within the 20-30 age range compared to salaried customers. ˆ No_of_Dependents: ˆ Salary: ˆ Partner_salary: ˆ Total_salary: ˆ Price: Slightly more dependents in business. Median salary of both professionals are same. No notable dierence. Median salary of both professionals are same. Business professionals buy more number of less-priced cars than salaried professionals. Marital Status: ˆ Age: ˆ No_of_Dependents: ˆ Salary: ˆ Partner_salary: Married people are generally older. Married individuals have more dependents. Median salary is slightly higher for single people. Broader range for married individuals, many without partner salary. This indicates that a salary of 25,600 Rs was incorrectly entered by approximately 11.6 percentage customers, so we replaced these values with 0. ˆ Total_salary: ˆ Price: Married individuals have higher incomes. Married people buy more expensive cars. Education: ˆ Age: ˆ No_of_Dependents: ˆ Salary: ˆ Postgraduates tend to be slightly older. No major dierence. Post graduates have higher salaries. Partner_salary: Similar distributions by education level. A higher number of postgraduates have partners with no salary compared to graduates. ˆ Total_salary: ˆ Price: Postgraduates have higher household incomes. Surprisingly graduates buy more expensive cars. Personal Loan: ˆ Age: Loan holders are younger. 18 ˆ No_of_Dependents: Number of dependents are less with customers with a personal loan consid- ering the median. Salary: Slightly lower for loan holders. ˆ Partner_salary: No major dierences. ˆ Total_salary: Slightly lower total incomes for loan holders. ˆ Price: Loan holders buy less expensive cars. House Loan: ˆ Age: Older individuals doesn't hold house loans. ˆ No_of_Dependents: No major dierences. ˆ Salary: Lower salaries for loan holders. ˆ Partner_salary: Customers with higher partner salaries doesn't hold loan. ˆ Total_salary: Lower total incomes for house loan holders. ˆ Price: Loan holders buy cheaper cars. Partner Working: ˆ Age: No signicant dierence. ˆ No_of_Dependents: Customers with working partners have zero dependents. ˆ Salary: Slightly lower for individuals with working partners. ˆ Partner_salary: This indicates that a salary of 25,600 Rs was incorrectly entered by approximately ˆ 12.6 percentage customers, so we replaced these values with 0. Total_salary: Higher total incomes for working partners. ˆ Price: slightly more customers buy expensive cars without working partners . Make (Car Type - SUV, Sedan, Hatchback): ˆ Age: SUV buyers are older, Hatchback buyers younger. ˆ No_of_Dependents: Sedan buyers have less dependents. ˆ Salary: Higher for SUV buyers, lower for Hatchback buyers. ˆ Partner_salary: Customers with high partner salaries buy SUVs. Customers ˆ with zero partner salaries prefer Hatchback ˆ ˆ Total_salary: Higher for SUV buyers. Price: SUVs are pricier, Hatchbacks are cheaper. Key Insights ˆ ˆ Age: Older buyers favor expensive cars (SUVs), with higher incomes. Gender: Females earn more and buy pricier cars as the male customers in our dataset are more younger their median salaries are less as comapred to female. We need to add more number of olders males to get a better insight. ˆ ˆ ˆ ˆ ˆ ˆ Profession: Salaried professionals earn more and buy pricier cars. Marital Status: Married individuals have lower incomes , but buy more expensive cars. Education: Postgraduates have higher incomes, but graduates buy more expensive cars. Loans: Loan holders (personal/house) buy lower-priced cars. Partner Working: Working partners boost household incomes and car price. Make: SUVs are bought by wealthier, older individuals; Hatchbacks by younger, lower-income buyers. 19 Figure 23: Barchart of 'Gender' column Observations 1. 2. 3. Males prefer Sedans and Hatchbacks equally, with fewer choosing SUVs. Females slightly favor SUVs over Sedans, with low interest in Hatchbacks. Overall, males dominate the Sedan and Hatchback categories, while SUVs are more balanced across genders. Insights 1. 2. 3. Gender impacts car preferences: Males favor Sedans/Hatchbacks, females prefer SUVs. Marketing Strategy: Target males for Sedans/Hatchbacks and females for SUVs. Product Focus: Tailor promotions to each gender's preferences. For instance, oering promotions, features, and advertisements that cater to males for Sedans and Hatchbacks, and to females for SUVs. 4. Supply Balance: Align inventory with gender-based demand to maximize eciency. This approach will help Austo Motor Company optimize customer engagement and product oerings. * Profession vs Make Figure 24: Barchart of 'Profession' column 20 Observations 1. 2. Business owners favor Sedans and Hatchbacks equally, with less preference for SUVs. Salaried professionals prefer Sedans, followed by Hatchbacks, with moderate interest in SUVs. Salaried professionals have more interest in products of Austo Motor Company than business professionals. Insights 1. 2. Profession impacts car choice: Business prefers Sedans/Hatchbacks, while Salaried professionals lean toward Sedans. Marketing Strategy: Focus SUV ads on Salaried group and Sedan/Hatchback ads for both groups. * Marital_status vs Make Figure 25: Barchart of 'Marital_status' column Observations 1. 2. Married individuals prefer Sedans most, followed by Hatchbacks and SUVs. Single individuals show little interest in Sedans and SUVs, preferring Hatchbacks slightly more. Insights 1. 2. Marital status aects car preference: Married people prefer Sedans/Hatchbacks, while Singles favor Hatchbacks. Marketing Strategy: Focus Sedan/Hatchback ads on Married individuals; tailor Hatchback campaigns for Singles. * Education vs Make 21 Figure 26: Barchart of 'Price' column Observations and Insights * SUVs are the least popular among both graduates and postgraduates. * Sedans are the most popular among postgraduates, followed by hatchbacks. * Hatchbacks are the second most preferred car among both education levels, with higher preference among postgraduates. * Graduates generally prefer sedans more than hatchbacks and SUVs. In the data we have, the number of postgraduates is more as compared to graduates. Given the higher count of postgraduates, the marketing team could tailor campaigns to appeal to this educated segment. Impact on Business * * SUVs, as they are less popular. hatchbacks. For postgraduates, sedans can The marketing strategy should focus on promoting For graduates, emphasize sedans and higher engagement. * Personal_loan vs Make Figure 27: Barchart of 'Personal_loan' column 22 drive Observations and Insights * Sedans are the most popular among customers with personal loans, followed by backs. * SUVs are * hatch- the least preferred for both personal loan and non-loan customers. Among non-loan customers, sedans and hatchbacks show similar demand, both higher than SUVs. Impact on Business * * * SUV promotions across both customer segments. Focus on sedan promotion for customers with personal loans. Continue a balanced hatchback promotion strategy, emphasizing Increase loans for increased en- gagement. * House_loan vs Make Figure 28: Barchart of 'House_loan' column Observations and Insights * Sedans are the most popular among customers without house loans, followed by backs. * SUVs are * the least preferred among both house loan and non-house loan customers. Customers with house loans show lower overall demand for cars, with backs hatch- sedans and hatch- preferred. The dataset shows that the number of customers with house loans is signicantly lower. Impact on Business * Increase SUV promotions across both customer segments, especially for house loan cus- tomers. * * Focus on sedans and hatchbacks for non-house loan customers. Target house loan customers with a focused approach to boost * Partner_working vs Make 23 sedan and hatchback sales. Figure 29: Barchart of 'Partner_working' column Observations: * The chart displays preferences for SUVs, Sedans, and Hatchbacks based on the partner's working status. * Sedans are highly favored by customers with working partners, followed by Hatchbacks. * Customers without working partners show almost equal preference for Sedans and Hatch- backs. * SUVs are the least popular in both cases. Insights: * Sedans are in high demand among customers with working partners, possibly for their comfort and space. * Low SUV demand suggests a need to reconsider its promotion in marketing strategies. * Hatchbacks are consistently popular across both groups, indicating stable demand. * Make vs Price Figure 30: Barchart of 'Make' column Observations: * The chart shows the distribution of car types (SUV, Sedan, Hatchback) based on Price. * SUVs are primarily priced between 20,000 and 70,000, with many high-priced models. * Sedans have a broader price range from 20,000 to 60,000. * Hatchbacks are mainly concentrated in the 20,000 to 40,000 price range. 24 Insights: * SUVs are concentrated in the mid-price range, implying limited demand for high-end models. * Sedans appeal to a wide customer base across various price ranges. * Hatchbacks are popular in the lower price segment, appealing to budget-conscious buyers. 4.3 Key questions to ponder 4.3.1 Do men tend to prefer SUVs more compared to women? No Women prefer SUVs more as compared to men. 4.3.2 What is the likelihood of a salaried person buying a Sedan? Likelihood of a salaried person buying a Sedan is 44.196%. 4.3.3 What evidence or data supports Sheldon Cooper's claim that a salaried male is an easier target for a SUV sale over a Sedan sale? Figure 31: Barchart of 'Make' column for salaried male According to the data, salaried males show a stronger preference for purchasing sedans over SUVs. Sheldon Cooper's assertion that salaried males are more susceptible to SUV sales doesn't hold true based on this evidence. 25 4.3.4 How does the the amount spent on purchasing automobiles vary by gender? (a) Countplot (b) Boxplot Figure 32: 'Gender' vs 'Price' Observation and Insights Male vs Female Spending Patterns: * Males predominantly purchase vehicles priced between 20,000 and 40,000 units, with a noticeable decline in higher price ranges. * Females tend to spend more, as most purchases are concentrated between 40,000 and 60,000 units. Gender-wise Spending Distribution: * The boxplot shows that females typically spend more on vehicles, with a higher median and a broader range of prices. * Males exhibit lower spending overall, with several outliers on the higher side, indicating a few instances of luxury purchases. Business Insight: * Males are price-conscious and purchase lower-cost models more frequently. * Females are willing to invest in higher-priced models, suggesting the potential for marketing premium models to the female demographic. 26 4.3.5 How much money was spent on purchasing automobiles by individuals who took a personal loan? (a) Countplot (b) Boxplot Figure 33: 'Personal_loan' vs 'Price' Observation and Insights Individuals with Personal Loans: * Most purchases fall between 20,000 and 36,000 units, with fewer purchases in higher price brackets. * The distribution skews towards lower-priced vehicles, suggesting loan-takers buy less expensive models. Individuals without Personal Loans: * * Purchases are concentrated between 20,000 and 35,000 units. This group shows a broader range of spending, indicating exibility in purchasing higherpriced models. Boxplot Comparison: * The median spending * of individuals without loans is higher than that of loan-takers. Loan-takers exhibit a more constrained spending range, while non-loan takers show a wider variation, purchasing higher-end models more frequently. Business Insight: 27 * Non-loan customers tend to spend more, suggesting premium models can be marketed to them. * Loan-takers prefer budget vehicles, highlighting the importance of nancing options for lower-cost models. 4.3.6 How does having a working partner inuence the purchase of higher-priced cars? (a) Countplot (b) Boxplot Figure 34: 'Partner_working' vs 'Price' Observations 1. Histograms: with working partners, demand is more evenly distributed across price 30,000 Rs.. * For customers without working partners, demand also peaks around 30,000 Rs., but declines sharply after 50,000 Rs.. Boxplot: * For customers ranges, peaking around 2. * Median car prices are similar for both groups. * Customers without working partners exhibit slightly more price variability. Insights * The price range around 30,000 Rs. has the highest demand across all customers. 28 * Working partners may be linked to a broader range of aordability. * Marketing campaigns should emphasize the 30,000 Rs. price point, with tailored strate- gies for higher-end models aimed at customers with working partners. 4.4 Some extra intersting questions to ponder 4.4.1 How does the level of Education impact price of cars purchased? (a) Countplot (b) Boxplot Figure 35: 'Education' vs 'Price' Observations 1. Histograms: * Male Graduates show the highest car purchase demand around Rs.30,000, with a sharp decline afterward. * Female Graduates show the highest demand around Rs.50,000. * Male Postgraduates have concentrated demand around Rs.30,000, graduates. 29 similar to male * Female Postgraduates show a broader demand distribution, peaking around Rs.45,000 Rs.55,000. Boxplots: * Postgraduates (both and 2. genders) show higher variability in car prices compared to grad- uates. * The median price for ates. * Male Postgraduates Male Graduates is signicantly lower than for have lower median prices compared to Female Gradu- Female Postgraduates. Insights * Graduates (especially males) prefer cars around Rs.30,000, while Female Graduates opt for higher-priced cars. * Postgraduates display more variability in spending, especially Female Postgraduates, with higher median prices. * Marketing campaigns should emphasize aordability around Rs.30,000 tomers, while highlighting mid-range options for females and postgraduates. 30 for male cus- 4.4.2 How does the total salary of customers inuence the price range of dierent car models (SUV, Sedan, Hatchback)? (a) Make - 'Sedan' (b) Make - 'Hatchback' 32 (c) Make - 'SUV' Observations and Insights 1. SUV: salaries Rs. 60k-80k, * * Most buyers have * Clustered buyers have A smaller segment consists of buyers with 2. 70k. Sedan: 3. Hatchback: * Buyers mainly have 32k. Rs. 48k-53k. salaries > Rs. 100k, buying purchasing at salaries Rs. 60k-80k, salaries Rs. 55k-70k, purchasing at purchasing at Insights for Business * SUV: Target mid-to-high salary customers. * Sedan: Focus on mid-range pricing. * Hatchback: Market to budget-conscious buyers. 33 at Rs. 60k- Rs. 20k-25k. Rs. 24k-26k and Rs. 30k-