Loading Data

GitHub Repository: HarshKapadia2/car-details

Load Data

data <- read.csv("https://raw.githubusercontent.com/HarshKapadia2/car-details/main/data/car_details_v3.csv")
head(data)

##                            name year selling_price km_driven   fuel seller_type
## 1        Maruti Swift Dzire VDI 2014        450000    145500 Diesel  Individual
## 2  Skoda Rapid 1.5 TDI Ambition 2014        370000    120000 Diesel  Individual
## 3      Honda City 2017-2020 EXi 2006        158000    140000 Petrol  Individual
## 4     Hyundai i20 Sportz Diesel 2010        225000    127000 Diesel  Individual
## 5        Maruti Swift VXI BSIII 2007        130000    120000 Petrol  Individual
## 6 Hyundai Xcent 1.2 VTVT E Plus 2017        440000     45000 Petrol  Individual
##   transmission        owner    mileage  engine  max_power
## 1       Manual  First Owner  23.4 kmpl 1248 CC     74 bhp
## 2       Manual Second Owner 21.14 kmpl 1498 CC 103.52 bhp
## 3       Manual  Third Owner  17.7 kmpl 1497 CC     78 bhp
## 4       Manual  First Owner  23.0 kmpl 1396 CC     90 bhp
## 5       Manual  First Owner  16.1 kmpl 1298 CC   88.2 bhp
## 6       Manual  First Owner 20.14 kmpl 1197 CC  81.86 bhp
##                     torque seats
## 1           190Nm@ 2000rpm     5
## 2      250Nm@ 1500-2500rpm     5
## 3    12.7@ 2,700(kgm@ rpm)     5
## 4 22.4 kgm at 1750-2750rpm     5
## 5    11.5@ 4,500(kgm@ rpm)     5
## 6        113.75nm@ 4000rpm     5

Structure of Data

str(data)

## 'data.frame':    8128 obs. of  13 variables:
##  $ name         : chr  "Maruti Swift Dzire VDI" "Skoda Rapid 1.5 TDI Ambition" "Honda City 2017-2020 EXi" "Hyundai i20 Sportz Diesel" ...
##  $ year         : int  2014 2014 2006 2010 2007 2017 2007 2001 2011 2013 ...
##  $ selling_price: int  450000 370000 158000 225000 130000 440000 96000 45000 350000 200000 ...
##  $ km_driven    : int  145500 120000 140000 127000 120000 45000 175000 5000 90000 169000 ...
##  $ fuel         : chr  "Diesel" "Diesel" "Petrol" "Diesel" ...
##  $ seller_type  : chr  "Individual" "Individual" "Individual" "Individual" ...
##  $ transmission : chr  "Manual" "Manual" "Manual" "Manual" ...
##  $ owner        : chr  "First Owner" "Second Owner" "Third Owner" "First Owner" ...
##  $ mileage      : chr  "23.4 kmpl" "21.14 kmpl" "17.7 kmpl" "23.0 kmpl" ...
##  $ engine       : chr  "1248 CC" "1498 CC" "1497 CC" "1396 CC" ...
##  $ max_power    : chr  "74 bhp" "103.52 bhp" "78 bhp" "90 bhp" ...
##  $ torque       : chr  "190Nm@ 2000rpm" "250Nm@ 1500-2500rpm" "12.7@ 2,700(kgm@ rpm)" "22.4 kgm at 1750-2750rpm" ...
##  $ seats        : int  5 5 5 5 5 5 5 4 5 5 ...

Installing Packages

# install.packages("tidyverse", repos = "http://cran.us.r-project.org") # For `ggplot2` and `dplyr`
install.packages("ggplot2", repos = "http://cran.us.r-project.org")

## Installing package into '/home/runner/work/_temp/Library'
## (as 'lib' is unspecified)

## also installing the dependencies 'colorspace', 'cli', 'crayon', 'utf8', 'farver', 'labeling', 'lifecycle', 'munsell', 'RColorBrewer', 'viridisLite', 'ellipsis', 'fansi', 'pillar', 'pkgconfig', 'vctrs', 'gtable', 'isoband', 'scales', 'tibble', 'withr'

install.packages("dplyr", repos = "http://cran.us.r-project.org")

## Installing package into '/home/runner/work/_temp/Library'
## (as 'lib' is unspecified)

## also installing the dependencies 'purrr', 'generics', 'tidyselect'

# install.packages("fansi", repos = "http://cran.us.r-project.org") # To solve `tidyverse` package error
install.packages("plotrix", repos = "http://cran.us.r-project.org") # For 3d Pie Chart

## Installing package into '/home/runner/work/_temp/Library'
## (as 'lib' is unspecified)

library(ggplot2)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(plotrix)
# library(tidyverse)

Data Cleaning and Processing

Remove Incomplete Rows

count(data)

##      n
## 1 8128

sum(is.na(data))

## [1] 221

row_status <- complete.cases(data)
data <- data[row_status,] # [row, column]
sum(is.na(data))

## [1] 0

count(data)

##      n
## 1 7907

Convert Engine Column from Character to Integer Vector

str(data$engine)

##  chr [1:7907] "1248 CC" "1498 CC" "1497 CC" "1396 CC" "1298 CC" "1197 CC" ...

engine_chr_values <- strsplit(data$engine, split = " ", fixed = TRUE)
engine_count <- length(engine_chr_values)
new_engine_values <- vector()

for (i in 1:engine_count) {
  new_engine_values[i] <- strtoi(engine_chr_values[[i]][1])
}

data$engine <- new_engine_values
str(data$engine)

##  int [1:7907] 1248 1498 1497 1396 1298 1197 1061 796 1364 1399 ...

Convert Mileage Column from Character to Numeric Vector

str(data$mileage)

##  chr [1:7907] "23.4 kmpl" "21.14 kmpl" "17.7 kmpl" "23.0 kmpl" "16.1 kmpl" ...

mileage_chr_values <- strsplit(data$mileage, split = " ", fixed = TRUE)
mileage_count <- length(mileage_chr_values)
new_mileage_values <- vector()

for (i in 1:mileage_count) {
  new_mileage_values[i] <- as.numeric(mileage_chr_values[[i]][1])
}

data$mileage <- new_mileage_values
str(data$mileage)

##  num [1:7907] 23.4 21.1 17.7 23 16.1 ...

Convert Max Power Column from Character to Numeric Vector

str(data$max_power)

##  chr [1:7907] "74 bhp" "103.52 bhp" "78 bhp" "90 bhp" "88.2 bhp" ...

max_power_chr_values <- strsplit(data$max_power, split = " ", fixed = TRUE)
max_power_count <- length(max_power_chr_values)
new_max_power_values <- vector()

for (i in 1:max_power_count) {
  new_max_power_values[i] <- as.numeric(max_power_chr_values[[i]][1])
}

data$max_power <- new_max_power_values
str(data$max_power)

##  num [1:7907] 74 103.5 78 90 88.2 ...

Data for Prediction

Select Numeric Data

pred_data <- data %>% select(selling_price, km_driven, owner, year, fuel, seller_type, transmission, mileage, engine)

Calculate Age

year_count <- length(pred_data$year)
age <- vector()

for (i in 1:year_count) {
  age[i] <- 2020 - pred_data$year[i]
}

pred_data <- cbind(pred_data, age)
str(pred_data$age)

##  num [1:7907] 6 6 14 10 13 3 13 19 9 7 ...

pred_data <- subset(pred_data, select = -year)
str(pred_data)

## 'data.frame':    7907 obs. of  9 variables:
##  $ selling_price: int  450000 370000 158000 225000 130000 440000 96000 45000 350000 200000 ...
##  $ km_driven    : int  145500 120000 140000 127000 120000 45000 175000 5000 90000 169000 ...
##  $ owner        : chr  "First Owner" "Second Owner" "Third Owner" "First Owner" ...
##  $ fuel         : chr  "Diesel" "Diesel" "Petrol" "Diesel" ...
##  $ seller_type  : chr  "Individual" "Individual" "Individual" "Individual" ...
##  $ transmission : chr  "Manual" "Manual" "Manual" "Manual" ...
##  $ mileage      : num  23.4 21.1 17.7 23 16.1 ...
##  $ engine       : int  1248 1498 1497 1396 1298 1197 1061 796 1364 1399 ...
##  $ age          : num  6 6 14 10 13 3 13 19 9 7 ...

Convert Fuel Types to Binary

1 = Petrol

0 = Diesel

pred_data %>% count(fuel)

##     fuel    n
## 1    CNG   53
## 2 Diesel 4299
## 3    LPG   35
## 4 Petrol 3520

pred_fuel_count <- length(pred_data$fuel)

for (i in 1:pred_fuel_count) {
  temp_val <- trimws(pred_data$fuel[i])

  if(temp_val %in% "CNG") {
    pred_data <- pred_data[-c(i),]
  }
  else if(temp_val %in% "LPG") {
    pred_data <- pred_data[-c(i),]
  }
}

pred_data %>% count(fuel)

##     fuel    n
## 1    CNG    2
## 2 Diesel 4299
## 3    LPG    1
## 4 Petrol 3520

fuel_type <- vector() # 1 = Petrol, 0 = Diesel
pred_fuel_count <- length(pred_data$fuel)

for(i in 1:pred_fuel_count) {
  if(pred_data$fuel[i] %in% "Petrol") {
    fuel_type[i] <- 1
  }
  else {
    fuel_type[i] <- 0
  }
}

pred_data <- cbind(pred_data, fuel_type)
pred_data <- subset(pred_data, select = -fuel)

str(pred_data)

## 'data.frame':    7822 obs. of  9 variables:
##  $ selling_price: int  450000 370000 158000 225000 130000 440000 45000 350000 200000 500000 ...
##  $ km_driven    : int  145500 120000 140000 127000 120000 45000 5000 90000 169000 68000 ...
##  $ owner        : chr  "First Owner" "Second Owner" "Third Owner" "First Owner" ...
##  $ seller_type  : chr  "Individual" "Individual" "Individual" "Individual" ...
##  $ transmission : chr  "Manual" "Manual" "Manual" "Manual" ...
##  $ mileage      : num  23.4 21.1 17.7 23 16.1 ...
##  $ engine       : int  1248 1498 1497 1396 1298 1197 796 1364 1399 1461 ...
##  $ age          : num  6 6 14 10 13 3 19 9 7 6 ...
##  $ fuel_type    : num  0 0 1 0 1 1 1 0 0 0 ...

Convert Transmission Types to Binary

1 = Manual

0 = Automatic

pred_data %>% count(transmission)

##   transmission    n
## 1    Automatic 1041
## 2       Manual 6781

transmission_type <- vector() # 1 = Manual, 0 = Automatic
transmission_count <- length(pred_data$transmission)

for(i in 1:transmission_count) {
  if(pred_data$transmission[i] %in% "Manual") {
    transmission_type[i] <- 1
  }
  else {
    transmission_type[i] <- 0
  }
}

pred_data <- cbind(pred_data, transmission_type)
pred_data <- subset(pred_data, select = -transmission)

str(pred_data)

## 'data.frame':    7822 obs. of  9 variables:
##  $ selling_price    : int  450000 370000 158000 225000 130000 440000 45000 350000 200000 500000 ...
##  $ km_driven        : int  145500 120000 140000 127000 120000 45000 5000 90000 169000 68000 ...
##  $ owner            : chr  "First Owner" "Second Owner" "Third Owner" "First Owner" ...
##  $ seller_type      : chr  "Individual" "Individual" "Individual" "Individual" ...
##  $ mileage          : num  23.4 21.1 17.7 23 16.1 ...
##  $ engine           : int  1248 1498 1497 1396 1298 1197 796 1364 1399 1461 ...
##  $ age              : num  6 6 14 10 13 3 19 9 7 6 ...
##  $ fuel_type        : num  0 0 1 0 1 1 1 0 0 0 ...
##  $ transmission_type: num  1 1 1 1 1 1 1 1 1 1 ...

Convert Seller Types to Binary

1 = Individual

0 = Dealer

pred_data %>% count(seller_type)

##        seller_type    n
## 1           Dealer 1106
## 2       Individual 6480
## 3 Trustmark Dealer  236

seller_count <- length(pred_data$seller_type)

for (i in 1:seller_count) {
  temp_val <- trimws(pred_data$seller_type[i])

  if(temp_val %in% "Trustmark Dealer") {
    pred_data <- pred_data[-c(i),]
  }
}

pred_data %>% count(seller_type)

##        seller_type    n
## 1           Dealer 1106
## 2       Individual 6480
## 3 Trustmark Dealer  100

seller_type <- vector() # 1 = Individual, 0 = Dealer
seller_count <- length(pred_data$seller_type)

for(i in 1:seller_count) {
  if(pred_data$seller_type[i] %in% "Individual") {
    seller_type[i] <- 1
  }
  else {
    seller_type[i] <- 0
  }
}

pred_data <- subset(pred_data, select = -seller_type)
pred_data <- cbind(pred_data, seller_type)

str(pred_data)

## 'data.frame':    7686 obs. of  9 variables:
##  $ selling_price    : int  450000 370000 158000 225000 130000 440000 45000 350000 200000 500000 ...
##  $ km_driven        : int  145500 120000 140000 127000 120000 45000 5000 90000 169000 68000 ...
##  $ owner            : chr  "First Owner" "Second Owner" "Third Owner" "First Owner" ...
##  $ mileage          : num  23.4 21.1 17.7 23 16.1 ...
##  $ engine           : int  1248 1498 1497 1396 1298 1197 796 1364 1399 1461 ...
##  $ age              : num  6 6 14 10 13 3 19 9 7 6 ...
##  $ fuel_type        : num  0 0 1 0 1 1 1 0 0 0 ...
##  $ transmission_type: num  1 1 1 1 1 1 1 1 1 1 ...
##  $ seller_type      : num  1 1 1 1 1 1 1 1 1 1 ...

Convert Owner Types to Binary

1 = First Owner

0 = Second Owner

pred_data %>% count(owner)

##                  owner    n
## 1          First Owner 5034
## 2 Fourth & Above Owner  158
## 3         Second Owner 1984
## 4       Test Drive Car    5
## 5          Third Owner  505

owner_count <- length(pred_data$owner)

for (i in 1:owner_count) {
  temp_val <- trimws(pred_data$owner[i])

  if(temp_val %in% "Fourth & Above Owner") {
    pred_data <- pred_data[-c(i),]
  }
  else if(temp_val %in% "Third Owner") {
    pred_data <- pred_data[-c(i),]
  }
  else if(temp_val %in% "Test Drive Car") {
    pred_data <- pred_data[-c(i),]
  }
}

pred_data %>% count(owner)

##                  owner    n
## 1          First Owner 5034
## 2 Fourth & Above Owner   22
## 3         Second Owner 1984
## 4       Test Drive Car    1
## 5          Third Owner   54

owner_type <- vector() # 1 = First Owner, 0 = Second Owner
owner_count <- length(pred_data$owner)

for(i in 1:owner_count) {
  if(pred_data$owner[i] %in% "First Owner") {
    owner_type[i] <- 1
  }
  else {
    owner_type[i] <- 0
  }
}

pred_data <- cbind(pred_data, owner_type)
pred_data <- subset(pred_data, select = -owner)

str(pred_data)

## 'data.frame':    7095 obs. of  9 variables:
##  $ selling_price    : int  450000 370000 225000 130000 440000 45000 350000 200000 500000 92000 ...
##  $ km_driven        : int  145500 120000 127000 120000 45000 5000 90000 169000 68000 100000 ...
##  $ mileage          : num  23.4 21.1 23 16.1 20.1 ...
##  $ engine           : int  1248 1498 1396 1298 1197 796 1364 1399 1461 993 ...
##  $ age              : num  6 6 10 13 3 19 9 7 6 15 ...
##  $ fuel_type        : num  0 0 0 1 1 1 0 0 0 1 ...
##  $ transmission_type: num  1 1 1 1 1 1 1 1 1 1 ...
##  $ seller_type      : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ owner_type       : num  1 0 1 1 1 0 1 1 0 0 ...

Visualising Data

Histogram

hist(data$seats, main = "No. of Seats vs No. of Cars", xlab = "No. of Seats", ylab = "No. of Cars")

Box Plot

temp_data <- data %>%
  select(selling_price, seats) %>%
  filter(selling_price < 200000)

boxplot(temp_data$selling_price ~ temp_data$seats, main = "No. of Seats vs Selling Price", xlab = "No. of Seats", ylab = "Selling Price (Rs)")

Pie Chart

fuel_count = data %>% count(fuel)
pie(fuel_count$n, labels = fuel_count$fuel, radius = 1, col = c("orange", "blue", "yellow", "dark green"), main = "Fuel Type")

3D Pie Chart

transmission_count = data %>% count(transmission)
pie3D(transmission_count$n, labels = transmission_count$transmission, radius = 1, col = c("yellow", "blue"), explode = 0.4, main = "Transmission Type")

Bar Plot

barplot(table(data$seller_type), main = "Seller Type vs No. of Cars")

Scatter Plot

graph_data_1 <- data %>%
  select(selling_price, km_driven) %>%
  filter(selling_price < 5000000, km_driven < 400000)

ggplot(graph_data_1, aes(x = selling_price, y = km_driven)) + geom_point() + geom_smooth() + labs(title = "Selling Price vs Distance Driven", x = "Selling Price (Rs)", y = "Distance Driven (km)")

## `geom_smooth()` using method = 'gam' and formula 'y ~ s(x, bs = "cs")'

Line Plot

graph_data_2 <- data %>%
  select(selling_price, km_driven) %>%
  filter(selling_price > 2000000, km_driven < 400000)

ggplot(graph_data_2, aes(x = selling_price, y = km_driven)) +
  stat_summary(fun.y = "mean", geom = "line", size = 1, linetype = "solid") +
  labs(title = "Selling Price vs Distance Driven", x = "Selling Price (Rs)", y = "Distance Driven (km)")

## Warning: `fun.y` is deprecated. Use `fun` instead.

Bar Chart

ggplot(data, aes(x = fuel, fill = owner)) + geom_bar() + labs(title = "Fuel Type vs No. of Cars vs Owner", x = "Fuel Type", y = "No. of Cars", fill = "Owner")

Dot Plot

ggplot(data, aes(x = seats, y = fuel)) + geom_point(col = "red", size = 5) +
  geom_segment(aes(x = min(seats), xend = max(seats), y = fuel, yend = fuel), linetype = "dashed", size = 0.1) +
  labs(title = "No. of Seats vs Fuel Type", x = "No. of Seats", y = "Fuel Type")

Smooth Plot

graph_data <- data %>%
  select(mileage, engine)

ggplot(graph_data, aes(x = engine, y = mileage)) + geom_smooth() + labs(title = "Engine Capacity vs Car Mileage", x = "Engine Capacity (cc)", y = "Car Mileage (km/L)")

## `geom_smooth()` using method = 'gam' and formula 'y ~ s(x, bs = "cs")'

Density Plot

ggplot(data, aes(mileage)) + geom_density(aes(fill = fuel), alpha = 0.3) + labs(title = "Car Mileage vs Density vs Fuel Type", x = "Car Mileage (km/L)", y = "Density", fill = "Fuel Type")

Prediction

Simple Linear Regression

plot(pred_data$engine, pred_data$mileage, xlab = "Engine Capacity (cc)", ylab = "Mileage (km/L)")
abline(lm(mileage ~ engine, data = pred_data), col = "red", lwd = 3)
title(main = "Engine Capacity vs Mileage")

plot(pred_data$age, pred_data$mileage, xlab = "Age (yr)", ylab = "Mileage (km/L)")
abline(lm(mileage ~ age, data = pred_data), col = "red", lwd = 3)
title(main = "Age vs Mileage")

Multiple Linear Regression

mlr <- lm(selling_price ~ km_driven + owner_type + age + fuel_type + seller_type + transmission_type + mileage + engine, data = pred_data)
summary(mlr)

## 
## Call:
## lm(formula = selling_price ~ km_driven + owner_type + age + fuel_type + 
##     seller_type + transmission_type + mileage + engine, data = pred_data)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1649833  -182047     6554   146683  7708354 
## 
## Coefficients:
##                     Estimate Std. Error t value Pr(>|t|)    
## (Intercept)        9.551e+05  9.759e+04   9.787  < 2e-16 ***
## km_driven         -1.424e+00  1.358e-01 -10.485  < 2e-16 ***
## owner_type         3.294e+04  1.627e+04   2.024  0.04296 *  
## age               -4.802e+04  2.362e+03 -20.325  < 2e-16 ***
## fuel_type         -5.672e+04  1.925e+04  -2.946  0.00323 ** 
## seller_type       -2.952e+05  1.905e+04 -15.499  < 2e-16 ***
## transmission_type -9.839e+05  2.246e+04 -43.802  < 2e-16 ***
## mileage            1.483e+04  2.667e+03   5.562 2.76e-08 ***
## engine             6.193e+02  2.426e+01  25.523  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 553200 on 7086 degrees of freedom
## Multiple R-squared:  0.5647, Adjusted R-squared:  0.5642 
## F-statistic:  1149 on 8 and 7086 DF,  p-value: < 2.2e-16

Car Details

Harsh Kapadia

2022-03-26

Loading Data

Load Data

Structure of Data

Installing Packages

Data Cleaning and Processing

Remove Incomplete Rows

Convert Engine Column from Character to Integer Vector

Convert Mileage Column from Character to Numeric Vector

Convert Max Power Column from Character to Numeric Vector

Data for Prediction

Select Numeric Data

Calculate Age

Convert Fuel Types to Binary

Convert Transmission Types to Binary

Convert Seller Types to Binary

Convert Owner Types to Binary

Visualising Data

Histogram

Box Plot

Pie Chart

3D Pie Chart

Bar Plot

Scatter Plot

Line Plot

Bar Chart

Dot Plot

Smooth Plot

Density Plot

Prediction

Simple Linear Regression

Multiple Linear Regression