Exploratory Data Analysis

TURKEY CAR MARKET 2020

1. Introduction

1.1 Turkey Online Car Market 2020

The dataset used in this project belongs to a website in Turkey which makes online buying and selling of cars advertised in 2020. It is called Turkey Car Market 2020 and downloaded from Kaggle. The dataset contains information about features of cars which has been filled by sellers. The missing features of cars in the dataset are written as “Don’t Know”. In order to overcome the rows with missing values, a preprocessing of data is conducted and can be seen in Preprocessing Section for detailed explanations. This section also includes some accuracy checks of the data, and explanations od the variables.

1.2 Objectives

In this project, we will first investigate the data for preprocessing to improve its quality. Then we will perform an exploratory data analysis(EDA) by data manipulation and data visualization steps. The main purpose is to identify which variables affect the price mostly and come up with a conclusion for the relationship between variables. In addition to these, we will study on some models to forecast prices of cars with given features. The processes during the assignment can be listed as below:

1. Data Preprocessing
2. Data Manipulation
3. Data Visualization
4. Interactive Shiny App
5. Forecasting

1.3 Used Packages

The packages used during the project can be listed as below:

1. tidyverse
2. lubridate
3. knitr
4. tinytex
5. data table
6. shiny
7. scales
8. kableExtra

#Required packages
pti <- c("data.table", "tidyverse", "knitr", "tinytex", "scales", "kableExtra")
pti <- pti[!(pti %in% installed.packages())]
if(length(pti)>0){
    install.packages(pti)
}

library(tidyverse)
library(knitr)
library(tinytex)
library(data.table)
library(scales)
library(kableExtra)

2. Exploratory Data Analysis - EDA

Before making any further analysis, we need to import the data that we pre-processed the data.The pre-processing report file link is given above.

#data uploading
carmarket = readRDS(gzcon(url("https://github.com/pjournal/boun01g-data-mine-r-s/blob/gh-pages/Project/turkey_car_market_EDA?raw=true")))

2.1 Time Series Analyses

In this dataset, there are two variables which are related to time, i.e., Model Year and Advertisement Date. To analyze of the variable relationships, we first present time variables analyses as time series analyses. In the description of the dataset, it is told that this dataset contains the advertisements in 2020 but there are two rows which are from 2019. We will consider that it is not a problem.

carmarket %>%
  group_by(Year) %>%
  summarise(count = n()) %>%
  arrange(Year) %>%
  select(Year, count)%>%
  kable(col.names = c("Advertisement Year", "Count")) %>%
  kable_minimal(full_width = F)

Advertisement Year	Count
2019	2
2020	8832

When we look at the bar plot, there are advertisement whose month is changing from march to June and also there are two advertisements from December of 2019. There are more advertisements in April respect to other months. We can look for the daily count of these advertisements.

carmarket %>%
  group_by(Month) %>%
  summarise(count = n()) %>%
  ggplot(., aes(x = as.factor(Month), y = count, fill=count)) +
  geom_col() +
  scale_fill_gradient("count", low="pink2", high="pink4") +
  theme_minimal() +
  theme(legend.position = "none", plot.title = element_text(vjust = 0.5)) +
  labs(title = "Number of Advertisements Over Months",
       subtitle = st,
       x = "Month",
       y = "Number of Advertisement")

carmarket %>%
  group_by(Date) %>%
  summarise(count = n()) %>%
  ggplot(., aes(x = Date, y = count)) +
  geom_line() +
  theme_minimal() +
  labs(title = "Number of Advertisements Over Days",
       subtitle = st,
       x = "Day",
       y = "Number of Advertisement")

There is a big jump in two days of April. So, we can get that with these commands:

carmarket %>%
  group_by(Date) %>%
  summarise(count = n()) %>%
  arrange(desc(count)) %>%
  select(Date, count)%>%
  kable(col.names = c("Date", "Count")) %>%
  kable_styling("striped", full_width = T) %>%
  scroll_box(width = "100%", height = "400px")

Date	Count
2020-04-18	3423
2020-04-17	1090
2020-05-12	353
2020-05-27	243
2020-04-22	149
2020-06-13	149
2020-06-12	139
2020-06-09	137
2020-05-01	136
2020-06-11	133
2020-06-16	124
2020-06-14	121
2020-06-15	115
2020-05-17	108
2020-06-08	89
2020-06-03	87
2020-06-02	76
2020-06-10	73
2020-06-01	68
2020-05-29	64
2020-06-05	62
2020-03-25	61
2020-06-06	59
2020-05-20	54
2020-05-28	53
2020-04-08	51
2020-05-30	51
2020-05-21	50
2020-06-07	47
2020-05-16	45
2020-06-04	43
2020-05-18	42
2020-05-13	41
2020-04-09	39
2020-05-09	37
2020-05-14	37
2020-05-06	36
2020-05-15	36
2020-04-02	34
2020-05-08	32
2020-05-11	32
2020-05-05	30
2020-05-22	30
2020-03-27	29
2020-04-01	29
2020-04-07	29
2020-05-19	29
2020-05-31	29
2020-04-30	28
2020-05-02	28
2020-05-03	28
2020-05-07	28
2020-04-29	26
2020-03-24	25
2020-03-26	25
2020-05-24	25
2020-03-19	24
2020-04-23	24
2020-04-13	23
2020-04-16	23
2020-04-20	23
2020-03-20	21
2020-04-10	21
2020-04-21	21
2020-04-24	21
2020-05-10	21
2020-05-23	21
2020-03-21	20
2020-03-22	20
2020-04-03	20
2020-04-27	20
2020-05-04	20
2020-03-28	19
2020-05-25	19
2020-05-26	19
2020-04-19	17
2020-04-28	17
2020-04-06	16
2020-04-15	16
2020-04-25	16
2020-04-11	15
2020-03-23	14
2020-03-29	13
2020-04-04	12
2020-04-05	12
2020-04-26	12
2020-03-30	11
2020-04-14	11
2020-04-12	10
2020-03-31	3
2019-12-23	1
2019-12-24	1

When we search for that increase in 17 - 18.04.2020 dates, there was an information about the sales that march, april and may have most of sales in a year (You can find this information from this link). So, this pick is an expected pick.

We can make a plot of sales respect to Brand.

carmarket  %>%
  group_by(Date, Brand) %>%
  summarise(count = n()) %>%
  ggplot(., aes(x = Date, y = count, color = Brand)) +
  geom_line() +
  theme_minimal() +
  labs(title = "Spread of Advertisement According to Date",
       subtitle = st,
       x = "Date",
       y = "Number of Advertisements",
       color = "Brands")

In the pick period we see that there is a type of brand that is mostly in the advertisement. We can inspect which brand it is. Note that, almost all advertisements accumulate between March and July. For this reason, to provide clear insight, we filter this time interval.

carmarket %>%
  group_by(Date, Brand) %>%
  summarise(count = n()) %>%
  arrange(desc(count)) %>%
  #head(10) %>%
  select(Date, Brand, count)%>%
  kable(col.names = c("Date", "Brand", "Count")) %>%
  kable_styling("striped", full_width = T) %>%
  scroll_box(width = "100%", height = "400px")

Date	Brand	Count
2020-04-18	Renault	1303
2020-04-18	BMW	241
2020-04-18	Mercedes	201
2020-04-18	Ford	183
2020-04-18	Hyundai	182
2020-04-18	Fiat	180
2020-04-18	Opel	178
2020-04-18	Audi	147
2020-04-18	Peugeot	130
2020-04-17	BMW	125
2020-05-27	Dacia	121
2020-04-17	Mercedes	117
2020-04-17	Audi	101
2020-04-17	Fiat	98
2020-04-17	Opel	93
2020-04-18	Citroen	92
2020-04-18	Skoda	82
2020-04-18	Dacia	81
2020-04-18	Nissan	81
2020-04-17	Ford	78
2020-04-17	Hyundai	76
2020-04-18	Honda	74
2020-04-17	Peugeot	73
2020-05-12	Renault	70
2020-04-18	Kia	65
2020-04-18	Seat	62
2020-06-09	Renault	53
2020-04-17	Skoda	51
2020-05-12	Mercedes	51
2020-04-18	Land Rover	50
2020-04-17	Nissan	45
2020-04-17	Land Rover	44
2020-04-22	Renault	43
2020-04-17	Dacia	38
2020-06-08	Renault	38
2020-05-12	Jeep	36
2020-04-18	Chevrolet	35
2020-03-25	Hyundai	32
2020-06-12	Renault	32
2020-05-12	BMW	31
2020-06-14	Renault	31
2020-04-17	Honda	30
2020-06-03	Hyundai	30
2020-06-13	Renault	30
2020-05-20	BMW	29
2020-06-16	Honda	29
2020-04-17	Kia	28
2020-06-02	Renault	28
2020-06-11	Peugeot	28
2020-06-11	Renault	28
2020-06-15	Renault	28
2020-04-17	Citroen	27
2020-05-01	Renault	27
2020-04-17	Seat	26
2020-05-21	Land Rover	26
2020-05-27	BMW	25
2020-05-27	Audi	24
2020-05-12	Hyundai	23
2020-05-27	Mercedes	23
2020-04-09	Audi	22
2020-05-29	Renault	21
2020-06-12	Opel	21
2020-06-13	Opel	21
2020-06-16	Renault	21
2020-05-27	Jaguar	20
2020-04-08	Jeep	19
2020-05-12	Fiat	19
2020-05-17	Renault	19
2020-05-12	Land Rover	18
2020-06-09	Fiat	18
2020-06-10	Renault	18
2020-06-13	Volkswagen	18
2020-04-17	Chevrolet	17
2020-04-18	Porsche	17
2020-06-01	Renault	17
2020-06-11	Hyundai	17
2020-05-01	Mercedes	16
2020-06-13	Fiat	16
2020-06-15	Opel	16
2020-04-22	Peugeot	15
2020-05-12	Ford	15
2020-06-14	Hyundai	15
2020-06-15	Fiat	14
2020-04-22	Hyundai	13
2020-05-01	Opel	13
2020-05-12	Skoda	13
2020-06-05	Renault	13
2020-06-13	Hyundai	13
2020-04-22	Ford	12
2020-05-12	Audi	12
2020-05-12	Peugeot	12
2020-05-28	Renault	12
2020-06-03	Fiat	12
2020-06-03	Renault	12
2020-06-06	Renault	12
2020-06-08	Ford	12
2020-06-12	Hyundai	12
2020-06-15	Hyundai	12
2020-06-16	Opel	12
2020-04-18	Jeep	11
2020-04-22	Mercedes	11
2020-05-13	Renault	11
2020-05-22	Renault	11
2020-05-30	Renault	11
2020-06-05	Fiat	11
2020-06-09	Ford	11
2020-06-12	Fiat	11
2020-06-14	Opel	11
2020-03-24	Ford	10
2020-04-18	Mitsubishi	10
2020-05-01	BMW	10
2020-05-01	Fiat	10
2020-05-12	Nissan	10
2020-05-12	Opel	10
2020-05-17	Fiat	10
2020-05-17	Ford	10
2020-05-17	Hyundai	10
2020-06-09	Opel	10
2020-06-10	Opel	10
2020-06-11	Opel	10
2020-06-16	Fiat	10
2020-06-16	Ford	10
2020-04-22	Audi	9
2020-04-22	BMW	9
2020-04-22	Fiat	9
2020-04-29	Renault	9
2020-05-01	Peugeot	9
2020-05-10	Renault	9
2020-05-17	Audi	9
2020-05-19	Renault	9
2020-05-24	Renault	9
2020-06-07	Renault	9
2020-06-13	Ford	9
2020-06-14	Ford	9
2020-05-01	Audi	8
2020-05-01	Ford	8
2020-05-02	Renault	8
2020-05-05	Renault	8
2020-05-15	Fiat	8
2020-05-16	Renault	8
2020-05-17	Mercedes	8
2020-05-30	Ford	8
2020-06-02	Opel	8
2020-06-03	Ford	8
2020-06-05	Opel	8
2020-06-06	Opel	8
2020-06-09	Hyundai	8
2020-06-10	Hyundai	8
2020-06-11	Ford	8
2020-06-11	Tofas	8
2020-06-13	Tofas	8
2020-06-14	Fiat	8
2020-06-15	Ford	8
2020-03-25	Opel	7
2020-04-07	Opel	7
2020-04-17	Porsche	7
2020-05-06	Opel	7
2020-05-06	Renault	7
2020-05-27	Ford	7
2020-05-29	Ford	7
2020-06-01	Ford	7
2020-06-04	Renault	7
2020-06-06	Ford	7
2020-06-07	Hyundai	7
2020-06-09	Peugeot	7
2020-06-12	Honda	7
2020-06-14	Tofas	7
2020-06-16	Peugeot	7
2020-03-21	Ford	6
2020-04-02	Hyundai	6
2020-04-08	Opel	6
2020-04-18	Mazda	6
2020-04-22	Nissan	6
2020-04-27	Renault	6
2020-04-30	Fiat	6
2020-05-08	Renault	6
2020-05-09	Fiat	6
2020-05-09	Opel	6
2020-05-11	Renault	6
2020-05-16	Hyundai	6
2020-05-16	Opel	6
2020-05-17	Nissan	6
2020-05-18	Hyundai	6
2020-05-20	Renault	6
2020-05-28	BMW	6
2020-05-29	Fiat	6
2020-06-01	Audi	6
2020-06-01	Honda	6
2020-06-04	Opel	6
2020-06-05	Ford	6
2020-06-08	Fiat	6
2020-06-08	Opel	6
2020-06-10	Peugeot	6
2020-06-11	Fiat	6
2020-06-12	BMW	6
2020-06-12	Dacia	6
2020-06-13	BMW	6
2020-06-15	Tofas	6
2020-06-16	Hyundai	6
2020-03-20	Mercedes	5
2020-03-22	Opel	5
2020-03-25	Fiat	5
2020-03-26	Hyundai	5
2020-03-27	BMW	5
2020-04-01	Ford	5
2020-04-02	Fiat	5
2020-04-02	Ford	5
2020-04-07	Hyundai	5
2020-04-08	Ford	5
2020-04-10	Fiat	5
2020-04-10	Opel	5
2020-04-17	Jeep	5
2020-04-21	Fiat	5
2020-04-23	Renault	5
2020-05-01	Kia	5
2020-05-01	Land Rover	5
2020-05-03	Ford	5
2020-05-03	Opel	5
2020-05-03	Renault	5
2020-05-06	Ford	5
2020-05-07	Renault	5
2020-05-08	Fiat	5
2020-05-09	Renault	5
2020-05-11	Fiat	5
2020-05-12	Citroen	5
2020-05-12	Seat	5
2020-05-14	Renault	5
2020-05-14	Skoda	5
2020-05-15	Hyundai	5
2020-05-15	Renault	5
2020-05-16	Kia	5
2020-05-17	BMW	5
2020-05-17	Land Rover	5
2020-05-17	Opel	5
2020-05-17	Peugeot	5
2020-05-18	Opel	5
2020-05-18	Peugeot	5
2020-05-18	Renault	5
2020-05-21	Fiat	5
2020-05-25	Renault	5
2020-05-26	Fiat	5
2020-05-27	Renault	5
2020-05-28	Audi	5
2020-05-28	Hyundai	5
2020-05-29	Opel	5
2020-05-29	Skoda	5
2020-05-30	Audi	5
2020-05-30	Citroen	5
2020-06-01	Fiat	5
2020-06-03	Opel	5
2020-06-04	Ford	5
2020-06-05	Hyundai	5
2020-06-06	Fiat	5
2020-06-07	Ford	5
2020-06-08	Peugeot	5
2020-06-09	BMW	5
2020-06-10	Audi	5
2020-06-10	Fiat	5
2020-06-10	Ford	5
2020-06-10	Mercedes	5
2020-06-11	Honda	5
2020-06-12	Chevrolet	5
2020-06-12	Ford	5
2020-06-14	Mercedes	5
2020-06-14	Peugeot	5
2020-06-15	Audi	5
2020-06-15	BMW	5
2020-06-16	BMW	5
2020-03-19	Fiat	4
2020-03-25	Ford	4
2020-03-26	Ford	4
2020-03-27	Peugeot	4
2020-03-28	Kia	4
2020-03-29	Fiat	4
2020-04-01	Fiat	4
2020-04-01	Hyundai	4
2020-04-03	Opel	4
2020-04-04	Opel	4
2020-04-06	Opel	4
2020-04-08	Acura	4
2020-04-08	Fiat	4
2020-04-09	Ford	4
2020-04-13	BMW	4
2020-04-15	Acura	4
2020-04-16	Fiat	4
2020-04-18	Chrysler	4
2020-04-19	Renault	4
2020-04-20	Acura	4
2020-04-20	Renault	4
2020-04-21	Renault	4
2020-04-22	Land Rover	4
2020-04-22	Opel	4
2020-04-24	Fiat	4
2020-04-25	Fiat	4
2020-04-28	Opel	4
2020-04-29	Opel	4
2020-05-01	Hyundai	4
2020-05-01	Nissan	4
2020-05-01	Skoda	4
2020-05-02	Mercedes	4
2020-05-03	Fiat	4
2020-05-03	Hyundai	4
2020-05-05	Fiat	4
2020-05-07	Chevrolet	4
2020-05-07	Fiat	4
2020-05-08	Ford	4
2020-05-08	Opel	4
2020-05-11	Opel	4
2020-05-12	Dacia	4
2020-05-12	Honda	4
2020-05-13	Ford	4
2020-05-13	Hyundai	4
2020-05-14	Ford	4
2020-05-14	Hyundai	4
2020-05-14	Mercedes	4
2020-05-15	Opel	4
2020-05-16	BMW	4
2020-05-17	Skoda	4
2020-05-18	Citroen	4
2020-05-18	Dacia	4
2020-05-19	Fiat	4
2020-05-19	Opel	4
2020-05-19	Peugeot	4
2020-05-21	Ford	4
2020-05-23	Ford	4
2020-05-23	Opel	4
2020-05-28	Fiat	4
2020-05-28	Honda	4
2020-05-28	Opel	4
2020-05-29	Audi	4
2020-05-30	Honda	4
2020-05-30	Peugeot	4
2020-05-31	Fiat	4
2020-05-31	Hyundai	4
2020-05-31	Renault	4
2020-06-01	Hyundai	4
2020-06-01	Opel	4
2020-06-02	Fiat	4
2020-06-02	Ford	4
2020-06-02	Hyundai	4
2020-06-02	Kia	4
2020-06-03	Mercedes	4
2020-06-04	Chevrolet	4
2020-06-05	Nissan	4
2020-06-06	Hyundai	4
2020-06-06	Peugeot	4
2020-06-07	BMW	4
2020-06-07	Honda	4
2020-06-07	Opel	4
2020-06-08	Hyundai	4
2020-06-09	Dacia	4
2020-06-09	Nissan	4
2020-06-09	Tofas	4
2020-06-11	Skoda	4
2020-06-12	Audi	4
2020-06-12	Citroen	4
2020-06-12	Nissan	4
2020-06-12	Peugeot	4
2020-06-12	Skoda	4
2020-06-12	Tofas	4
2020-06-13	Honda	4
2020-06-13	Peugeot	4
2020-06-13	Skoda	4
2020-06-14	Chevrolet	4
2020-06-14	Skoda	4
2020-06-15	Nissan	4
2020-06-16	Citroen	4
2020-06-16	Dacia	4
2020-03-19	Ford	3
2020-03-21	Fiat	3
2020-03-23	Ford	3
2020-03-24	Fiat	3
2020-03-25	Peugeot	3
2020-03-26	Mercedes	3
2020-03-26	Opel	3
2020-03-27	Fiat	3
2020-03-27	Honda	3
2020-03-27	Hyundai	3
2020-03-27	Mercedes	3
2020-03-27	Opel	3
2020-03-28	Peugeot	3
2020-04-02	Opel	3
2020-04-02	Skoda	3
2020-04-03	Fiat	3
2020-04-04	Audi	3
2020-04-07	Fiat	3
2020-04-07	Ford	3
2020-04-09	Hyundai	3
2020-04-13	Honda	3
2020-04-14	Peugeot	3
2020-04-15	Ford	3
2020-04-16	Ford	3
2020-04-16	Opel	3
2020-04-18	Alfa Romeo	3
2020-04-18	Lada	3
2020-04-20	Opel	3
2020-04-22	Seat	3
2020-04-22	Skoda	3
2020-04-23	Ford	3
2020-04-24	Peugeot	3
2020-04-25	Ford	3
2020-04-25	Renault	3
2020-04-26	Renault	3
2020-04-28	Peugeot	3
2020-04-29	Fiat	3
2020-04-29	Peugeot	3
2020-04-30	Opel	3
2020-04-30	Porsche	3
2020-04-30	Renault	3
2020-05-01	Honda	3
2020-05-01	Seat	3
2020-05-02	Fiat	3
2020-05-03	Honda	3
2020-05-04	Ford	3
2020-05-04	Hyundai	3
2020-05-05	Ford	3
2020-05-05	Mercedes	3
2020-05-06	BMW	3
2020-05-07	Hyundai	3
2020-05-08	Hyundai	3
2020-05-08	Skoda	3
2020-05-09	Chevrolet	3
2020-05-09	Ford	3
2020-05-11	BMW	3
2020-05-11	Ford	3
2020-05-12	Kia	3
2020-05-12	Porsche	3
2020-05-13	Audi	3
2020-05-13	Mercedes	3
2020-05-14	Audi	3
2020-05-14	Honda	3
2020-05-17	Dacia	3
2020-05-17	Porsche	3
2020-05-18	Fiat	3
2020-05-20	Honda	3
2020-05-20	Opel	3
2020-05-21	Opel	3
2020-05-22	Dacia	3
2020-05-22	Fiat	3
2020-05-22	Hyundai	3
2020-05-23	Renault	3
2020-05-24	Hyundai	3
2020-05-24	Opel	3
2020-05-25	Hyundai	3
2020-05-26	Renault	3
2020-05-27	Fiat	3
2020-05-27	Opel	3
2020-05-27	Skoda	3
2020-05-28	Ford	3
2020-05-29	Hyundai	3
2020-05-29	Kia	3
2020-05-30	Hyundai	3
2020-05-30	Mercedes	3
2020-05-31	Honda	3
2020-05-31	Nissan	3
2020-05-31	Opel	3
2020-06-01	Citroen	3
2020-06-01	Peugeot	3
2020-06-01	Seat	3
2020-06-01	Skoda	3
2020-06-02	Chevrolet	3
2020-06-02	Citroen	3
2020-06-02	Honda	3
2020-06-02	Skoda	3
2020-06-03	Nissan	3
2020-06-04	Fiat	3
2020-06-04	Nissan	3
2020-06-06	Mercedes	3
2020-06-06	Seat	3
2020-06-07	Fiat	3
2020-06-07	Mercedes	3
2020-06-08	Nissan	3
2020-06-08	Skoda	3
2020-06-09	Mercedes	3
2020-06-11	Chevrolet	3
2020-06-11	Mercedes	3
2020-06-11	Seat	3
2020-06-12	Kia	3
2020-06-13	Dacia	3
2020-06-13	Mercedes	3
2020-06-13	Nissan	3
2020-06-14	Citroen	3
2020-06-14	Dacia	3
2020-06-14	Honda	3
2020-06-15	Skoda	3
2020-06-16	Mercedes	3
2020-06-16	Seat	3
2020-03-19	Dacia	2
2020-03-19	Honda	2
2020-03-19	Hyundai	2
2020-03-19	Mercedes	2
2020-03-19	Peugeot	2
2020-03-19	Seat	2
2020-03-20	BMW	2
2020-03-20	Chrysler	2
2020-03-20	Honda	2
2020-03-20	Hyundai	2
2020-03-21	Hyundai	2
2020-03-22	Audi	2
2020-03-22	Ford	2
2020-03-22	Nissan	2
2020-03-23	Fiat	2
2020-03-23	Hyundai	2
2020-03-23	Opel	2
2020-03-23	Peugeot	2
2020-03-24	Dacia	2
2020-03-24	Opel	2
2020-03-24	Peugeot	2
2020-03-25	Audi	2
2020-03-25	Nissan	2
2020-03-26	Fiat	2
2020-03-26	Peugeot	2
2020-03-28	Ford	2
2020-03-28	Opel	2
2020-03-29	Chevrolet	2
2020-03-29	Hyundai	2
2020-03-29	Opel	2
2020-03-30	Chevrolet	2
2020-03-30	Fiat	2
2020-03-30	Ford	2
2020-04-01	Audi	2
2020-04-01	BMW	2
2020-04-01	Kia	2
2020-04-01	Nissan	2
2020-04-01	Skoda	2
2020-04-02	Audi	2
2020-04-02	Citroen	2
2020-04-02	Mercedes	2
2020-04-03	Citroen	2
2020-04-03	Kia	2
2020-04-04	Peugeot	2
2020-04-05	Fiat	2
2020-04-05	Mercedes	2
2020-04-05	Opel	2
2020-04-06	Fiat	2
2020-04-06	Hyundai	2
2020-04-07	Dacia	2
2020-04-07	Honda	2
2020-04-07	Skoda	2
2020-04-08	BMW	2
2020-04-08	Hyundai	2
2020-04-08	Skoda	2
2020-04-09	Acura	2
2020-04-09	Fiat	2
2020-04-09	Mercedes	2
2020-04-10	BMW	2
2020-04-10	Ford	2
2020-04-10	Hyundai	2
2020-04-10	Mercedes	2
2020-04-11	Chevrolet	2
2020-04-11	Fiat	2
2020-04-11	Hyundai	2
2020-04-11	Opel	2
2020-04-12	Hyundai	2
2020-04-13	Fiat	2
2020-04-13	Ford	2
2020-04-13	Hyundai	2
2020-04-13	Nissan	2
2020-04-13	Opel	2
2020-04-14	Acura	2
2020-04-14	BMW	2
2020-04-15	Honda	2
2020-04-15	Hyundai	2
2020-04-15	Peugeot	2
2020-04-16	Honda	2
2020-04-16	Hyundai	2
2020-04-16	Mercedes	2
2020-04-16	Peugeot	2
2020-04-16	Skoda	2
2020-04-17	Alfa Romeo	2
2020-04-17	Infiniti	2
2020-04-17	Jaguar	2
2020-04-17	Mitsubishi	2
2020-04-18	Jaguar	2
2020-04-19	BMW	2
2020-04-19	Kia	2
2020-04-19	Nissan	2
2020-04-19	Opel	2
2020-04-20	Hyundai	2
2020-04-20	Jeep	2
2020-04-20	Skoda	2
2020-04-21	Ford	2
2020-04-21	Hyundai	2
2020-04-21	Land Rover	2
2020-04-21	Skoda	2
2020-04-22	Honda	2
2020-04-22	Kia	2
2020-04-23	Acura	2
2020-04-23	Citroen	2
2020-04-23	Fiat	2
2020-04-23	Hyundai	2
2020-04-24	BMW	2
2020-04-24	Ford	2
2020-04-24	Renault	2
2020-04-27	Acura	2
2020-04-27	Opel	2
2020-04-27	Peugeot	2
2020-04-28	Chevrolet	2
2020-04-28	Skoda	2
2020-04-29	Ford	2
2020-04-29	Hyundai	2
2020-04-30	Mercedes	2
2020-04-30	Seat	2
2020-04-30	Skoda	2
2020-05-01	Citroen	2
2020-05-01	Porsche	2
2020-05-02	BMW	2
2020-05-02	Dacia	2
2020-05-02	Ford	2
2020-05-02	Peugeot	2
2020-05-04	Fiat	2
2020-05-04	Peugeot	2
2020-05-04	Renault	2
2020-05-05	BMW	2
2020-05-05	Opel	2
2020-05-05	Peugeot	2
2020-05-06	Dacia	2
2020-05-06	Fiat	2
2020-05-06	Peugeot	2
2020-05-06	Skoda	2
2020-05-07	BMW	2
2020-05-07	Citroen	2
2020-05-07	Ford	2
2020-05-08	Mercedes	2
2020-05-09	Audi	2
2020-05-09	BMW	2
2020-05-09	Jeep	2
2020-05-11	Mercedes	2
2020-05-12	Alfa Romeo	2
2020-05-12	Chevrolet	2
2020-05-12	Jaguar	2
2020-05-12	Mazda	2
2020-05-13	Honda	2
2020-05-13	Opel	2
2020-05-13	Peugeot	2
2020-05-14	BMW	2
2020-05-14	Dacia	2
2020-05-14	Fiat	2
2020-05-15	Audi	2
2020-05-15	Ford	2
2020-05-15	Peugeot	2
2020-05-16	Chrysler	2
2020-05-16	Citroen	2
2020-05-16	Fiat	2
2020-05-16	Ford	2
2020-05-16	Mitsubishi	2
2020-05-16	Peugeot	2
2020-05-17	Seat	2
2020-05-18	Ford	2
2020-05-18	Honda	2
2020-05-18	Mercedes	2
2020-05-18	Skoda	2
2020-05-19	Ford	2
2020-05-19	Honda	2
2020-05-20	Citroen	2
2020-05-20	Fiat	2
2020-05-20	Mercedes	2
2020-05-21	Chevrolet	2
2020-05-21	Hyundai	2
2020-05-21	Renault	2
2020-05-22	Opel	2
2020-05-23	Audi	2
2020-05-23	Hyundai	2
2020-05-24	Nissan	2
2020-05-24	Peugeot	2
2020-05-25	Nissan	2
2020-05-26	Citroen	2
2020-05-26	Ford	2
2020-05-26	Honda	2
2020-05-26	Hyundai	2
2020-05-27	Honda	2
2020-05-27	Nissan	2
2020-05-27	Peugeot	2
2020-05-28	Nissan	2
2020-05-29	Acura	2
2020-05-29	BMW	2
2020-05-29	Mazda	2
2020-05-30	BMW	2
2020-05-31	Ford	2
2020-06-01	BMW	2
2020-06-01	Dacia	2
2020-06-01	Nissan	2
2020-06-02	Mercedes	2
2020-06-02	Mitsubishi	2
2020-06-02	Nissan	2
2020-06-02	Peugeot	2
2020-06-03	Citroen	2
2020-06-03	Dacia	2
2020-06-03	Peugeot	2
2020-06-03	Skoda	2
2020-06-04	BMW	2
2020-06-04	Dacia	2
2020-06-04	Honda	2
2020-06-04	Hyundai	2
2020-06-04	Mercedes	2
2020-06-04	Tofas	2
2020-06-05	Audi	2
2020-06-05	BMW	2
2020-06-05	Mercedes	2
2020-06-05	Tofas	2
2020-06-06	BMW	2
2020-06-06	Nissan	2
2020-06-06	Tofas	2
2020-06-07	Audi	2
2020-06-08	BMW	2
2020-06-08	Chevrolet	2
2020-06-08	Honda	2
2020-06-08	Mercedes	2
2020-06-09	Kia	2
2020-06-09	Mitsubishi	2
2020-06-10	BMW	2
2020-06-10	Honda	2
2020-06-10	Nissan	2
2020-06-11	Acura	2
2020-06-11	Dacia	2
2020-06-11	Nissan	2
2020-06-12	Mercedes	2
2020-06-13	Citroen	2
2020-06-13	Kia	2
2020-06-14	BMW	2
2020-06-14	Kia	2
2020-06-14	Mitsubishi	2
2020-06-14	Seat	2
2020-06-15	Chevrolet	2
2020-06-15	Dacia	2
2020-06-15	Peugeot	2
2020-06-16	Audi	2
2020-06-16	Nissan	2
2020-06-16	Skoda	2
2019-12-23	Hyundai	1
2019-12-24	Seat	1
2020-03-19	Audi	1
2020-03-19	BMW	1
2020-03-19	Chevrolet	1
2020-03-19	Citroen	1
2020-03-19	Opel	1
2020-03-20	Audi	1
2020-03-20	Chevrolet	1
2020-03-20	Fiat	1
2020-03-20	Ford	1
2020-03-20	Geely	1
2020-03-20	Opel	1
2020-03-20	Peugeot	1
2020-03-20	Skoda	1
2020-03-21	Audi	1
2020-03-21	BMW	1
2020-03-21	Chevrolet	1
2020-03-21	Dacia	1
2020-03-21	Honda	1
2020-03-21	Kia	1
2020-03-21	Nissan	1
2020-03-21	Opel	1
2020-03-21	Skoda	1
2020-03-22	Acura	1
2020-03-22	BMW	1
2020-03-22	Citroen	1
2020-03-22	Fiat	1
2020-03-22	Honda	1
2020-03-22	Hyundai	1
2020-03-22	Lada	1
2020-03-22	Peugeot	1
2020-03-22	Seat	1
2020-03-23	Audi	1
2020-03-23	Honda	1
2020-03-23	Seat	1
2020-03-24	Alfa Romeo	1
2020-03-24	BMW	1
2020-03-24	Citroen	1
2020-03-24	Hyundai	1
2020-03-24	Kia	1
2020-03-24	Land Rover	1
2020-03-25	BMW	1
2020-03-25	Honda	1
2020-03-25	Lada	1
2020-03-25	Mazda	1
2020-03-25	Mercedes	1
2020-03-25	Mitsubishi	1
2020-03-26	Acura	1
2020-03-26	Audi	1
2020-03-26	Citroen	1
2020-03-26	Honda	1
2020-03-26	Kia	1
2020-03-26	Land Rover	1
2020-03-27	Acura	1
2020-03-27	Audi	1
2020-03-27	Ford	1
2020-03-27	Kia	1
2020-03-27	Nissan	1
2020-03-28	Audi	1
2020-03-28	BMW	1
2020-03-28	Honda	1
2020-03-28	Land Rover	1
2020-03-28	Mercedes	1
2020-03-28	Nissan	1
2020-03-28	Seat	1
2020-03-28	Skoda	1
2020-03-29	Honda	1
2020-03-29	Mercedes	1
2020-03-29	Nissan	1
2020-03-30	Audi	1
2020-03-30	Hyundai	1
2020-03-30	Mercedes	1
2020-03-30	Opel	1
2020-03-30	Peugeot	1
2020-03-31	Fiat	1
2020-03-31	Ford	1
2020-03-31	Peugeot	1
2020-04-01	Acura	1
2020-04-01	Chevrolet	1
2020-04-01	Mercedes	1
2020-04-01	Opel	1
2020-04-01	Peugeot	1
2020-04-01	Seat	1
2020-04-02	Alfa Romeo	1
2020-04-02	Honda	1
2020-04-02	Jeep	1
2020-04-02	Kia	1
2020-04-02	Mazda	1
2020-04-02	Peugeot	1
2020-04-03	Acura	1
2020-04-03	BMW	1
2020-04-03	Chevrolet	1
2020-04-03	Dacia	1
2020-04-03	Ford	1
2020-04-03	Honda	1
2020-04-03	Land Rover	1
2020-04-03	Mazda	1
2020-04-03	Mitsubishi	1
2020-04-04	Fiat	1
2020-04-04	Honda	1
2020-04-04	Hyundai	1
2020-04-05	BMW	1
2020-04-05	Citroen	1
2020-04-05	Ford	1
2020-04-05	Hyundai	1
2020-04-05	Kia	1
2020-04-05	Nissan	1
2020-04-06	Acura	1
2020-04-06	BMW	1
2020-04-06	Chevrolet	1
2020-04-06	Ford	1
2020-04-06	Honda	1
2020-04-06	Peugeot	1
2020-04-06	Seat	1
2020-04-06	Skoda	1
2020-04-07	Chevrolet	1
2020-04-07	Lada	1
2020-04-07	Mercedes	1
2020-04-07	Peugeot	1
2020-04-07	Seat	1
2020-04-08	Chevrolet	1
2020-04-08	Honda	1
2020-04-08	Kia	1
2020-04-08	Land Rover	1
2020-04-08	Mercedes	1
2020-04-08	Peugeot	1
2020-04-08	Seat	1
2020-04-09	Dacia	1
2020-04-09	Kia	1
2020-04-09	Peugeot	1
2020-04-09	Skoda	1
2020-04-10	Audi	1
2020-04-10	Dacia	1
2020-04-10	Skoda	1
2020-04-11	BMW	1
2020-04-11	Ford	1
2020-04-11	Lada	1
2020-04-11	Mercedes	1
2020-04-11	Peugeot	1
2020-04-11	Seat	1
2020-04-11	Skoda	1
2020-04-12	Acura	1
2020-04-12	Chevrolet	1
2020-04-12	Fiat	1
2020-04-12	Honda	1
2020-04-12	Lada	1
2020-04-12	Mitsubishi	1
2020-04-12	Peugeot	1
2020-04-12	Skoda	1
2020-04-13	Citroen	1
2020-04-13	Kia	1
2020-04-13	Mercedes	1
2020-04-13	Peugeot	1
2020-04-13	Seat	1
2020-04-13	Skoda	1
2020-04-14	Audi	1
2020-04-14	Citroen	1
2020-04-14	Dacia	1
2020-04-14	Hyundai	1
2020-04-15	Fiat	1
2020-04-15	Mazda	1
2020-04-15	Mercedes	1
2020-04-16	Audi	1
2020-04-16	Dacia	1
2020-04-16	Kia	1
2020-04-17	Chrysler	1
2020-04-17	Lada	1
2020-04-17	Mazda	1
2020-04-19	Audi	1
2020-04-19	Fiat	1
2020-04-19	Honda	1
2020-04-19	Hyundai	1
2020-04-19	Skoda	1
2020-04-20	Audi	1
2020-04-20	BMW	1
2020-04-20	Citroen	1
2020-04-20	Ford	1
2020-04-20	Kia	1
2020-04-20	Peugeot	1
2020-04-21	Acura	1
2020-04-21	Citroen	1
2020-04-21	Opel	1
2020-04-21	Peugeot	1
2020-04-22	Chevrolet	1
2020-04-22	Citroen	1
2020-04-22	Dacia	1
2020-04-22	Jeep	1
2020-04-23	Audi	1
2020-04-23	BMW	1
2020-04-23	Chevrolet	1
2020-04-23	Dacia	1
2020-04-23	Land Rover	1
2020-04-23	Mercedes	1
2020-04-23	Nissan	1
2020-04-23	Opel	1
2020-04-24	Acura	1
2020-04-24	Audi	1
2020-04-24	Dacia	1
2020-04-24	Hyundai	1
2020-04-24	Land Rover	1
2020-04-24	Mazda	1
2020-04-24	Nissan	1
2020-04-24	Skoda	1
2020-04-25	Citroen	1
2020-04-25	Jeep	1
2020-04-25	Mercedes	1
2020-04-25	Nissan	1
2020-04-25	Opel	1
2020-04-25	Skoda	1
2020-04-26	Audi	1
2020-04-26	Chevrolet	1
2020-04-26	Citroen	1
2020-04-26	Fiat	1
2020-04-26	Ford	1
2020-04-26	Hyundai	1
2020-04-26	Opel	1
2020-04-26	Peugeot	1
2020-04-26	Seat	1
2020-04-27	Dacia	1
2020-04-27	Ford	1
2020-04-27	Honda	1
2020-04-27	Hyundai	1
2020-04-27	Land Rover	1
2020-04-27	Mercedes	1
2020-04-27	Seat	1
2020-04-27	Skoda	1
2020-04-28	BMW	1
2020-04-28	Citroen	1
2020-04-28	Honda	1
2020-04-28	Land Rover	1
2020-04-28	Renault	1
2020-04-28	Seat	1
2020-04-29	Dacia	1
2020-04-29	Mazda	1
2020-04-29	Skoda	1
2020-04-30	Chevrolet	1
2020-04-30	Ford	1
2020-04-30	Honda	1
2020-04-30	Hyundai	1
2020-04-30	Kia	1
2020-04-30	Nissan	1
2020-04-30	Peugeot	1
2020-05-01	Chevrolet	1
2020-05-01	Dacia	1
2020-05-01	Jeep	1
2020-05-02	Honda	1
2020-05-02	Hyundai	1
2020-05-02	Kia	1
2020-05-02	Opel	1
2020-05-02	Seat	1
2020-05-03	Land Rover	1
2020-05-03	Skoda	1
2020-05-04	BMW	1
2020-05-04	Dacia	1
2020-05-04	Honda	1
2020-05-04	Kia	1
2020-05-04	Land Rover	1
2020-05-04	Nissan	1
2020-05-04	Opel	1
2020-05-04	Skoda	1
2020-05-05	Audi	1
2020-05-05	Chevrolet	1
2020-05-05	Dacia	1
2020-05-05	Honda	1
2020-05-05	Hyundai	1
2020-05-05	Skoda	1
2020-05-06	Chevrolet	1
2020-05-06	Chrysler	1
2020-05-06	Honda	1
2020-05-06	Hyundai	1
2020-05-06	Mercedes	1
2020-05-06	Seat	1
2020-05-07	Acura	1
2020-05-07	Honda	1
2020-05-07	Jaguar	1
2020-05-07	Land Rover	1
2020-05-07	Peugeot	1
2020-05-07	Skoda	1
2020-05-08	Acura	1
2020-05-08	Audi	1
2020-05-08	Dacia	1
2020-05-08	Peugeot	1
2020-05-08	Seat	1
2020-05-09	Citroen	1
2020-05-09	Honda	1
2020-05-09	Hyundai	1
2020-05-09	Kia	1
2020-05-09	Mercedes	1
2020-05-09	Nissan	1
2020-05-09	Peugeot	1
2020-05-09	Skoda	1
2020-05-10	BMW	1
2020-05-10	Citroen	1
2020-05-10	Fiat	1
2020-05-10	Ford	1
2020-05-10	Honda	1
2020-05-10	Infiniti	1
2020-05-10	Mazda	1
2020-05-10	Mercedes	1
2020-05-10	Nissan	1
2020-05-10	Opel	1
2020-05-10	Porsche	1
2020-05-10	Skoda	1
2020-05-11	Audi	1
2020-05-11	Chevrolet	1
2020-05-11	Honda	1
2020-05-11	Hyundai	1
2020-05-11	Kia	1
2020-05-11	Mitsubishi	1
2020-05-11	Nissan	1
2020-05-11	Peugeot	1
2020-05-11	Skoda	1
2020-05-12	Mitsubishi	1
2020-05-13	BMW	1
2020-05-13	Chevrolet	1
2020-05-13	Dacia	1
2020-05-13	Fiat	1
2020-05-13	Geely	1
2020-05-13	Kia	1
2020-05-13	Land Rover	1
2020-05-13	Rover	1
2020-05-13	Seat	1
2020-05-13	Skoda	1
2020-05-14	Citroen	1
2020-05-14	Opel	1
2020-05-14	Peugeot	1
2020-05-15	Chevrolet	1
2020-05-15	Citroen	1
2020-05-15	Honda	1
2020-05-15	Kia	1
2020-05-15	Mazda	1
2020-05-15	Mercedes	1
2020-05-15	Seat	1
2020-05-15	Skoda	1
2020-05-16	Audi	1
2020-05-16	Chevrolet	1
2020-05-16	Mercedes	1
2020-05-16	Seat	1
2020-05-17	Alfa Romeo	1
2020-05-17	Chrysler	1
2020-05-17	Honda	1
2020-05-17	Kia	1
2020-05-18	BMW	1
2020-05-18	Seat	1
2020-05-19	BMW	1
2020-05-19	Chevrolet	1
2020-05-19	Hyundai	1
2020-05-19	Skoda	1
2020-05-20	Chevrolet	1
2020-05-20	Dacia	1
2020-05-20	Hyundai	1
2020-05-20	Kia	1
2020-05-20	Peugeot	1
2020-05-20	Seat	1
2020-05-20	Skoda	1
2020-05-21	Audi	1
2020-05-21	Citroen	1
2020-05-21	Dacia	1
2020-05-21	Mercedes	1
2020-05-21	Peugeot	1
2020-05-21	Skoda	1
2020-05-22	BMW	1
2020-05-22	Chevrolet	1
2020-05-22	Ford	1
2020-05-22	Honda	1
2020-05-22	Kia	1
2020-05-22	Land Rover	1
2020-05-22	Mercedes	1
2020-05-22	Skoda	1
2020-05-23	Dacia	1
2020-05-23	Fiat	1
2020-05-23	Honda	1
2020-05-23	Kia	1
2020-05-23	Peugeot	1
2020-05-23	Seat	1
2020-05-24	Acura	1
2020-05-24	BMW	1
2020-05-24	Chevrolet	1
2020-05-24	Citroen	1
2020-05-24	Fiat	1
2020-05-24	Lada	1
2020-05-25	Audi	1
2020-05-25	BMW	1
2020-05-25	Fiat	1
2020-05-25	Honda	1
2020-05-25	Kia	1
2020-05-25	Lada	1
2020-05-25	Mercedes	1
2020-05-25	Opel	1
2020-05-25	Peugeot	1
2020-05-26	Audi	1
2020-05-26	BMW	1
2020-05-26	Kia	1
2020-05-27	Kia	1
2020-05-27	Porsche	1
2020-05-27	Rover	1
2020-05-28	Alfa Romeo	1
2020-05-28	Chevrolet	1
2020-05-28	Citroen	1
2020-05-28	Dacia	1
2020-05-28	Land Rover	1
2020-05-28	Porsche	1
2020-05-28	Seat	1
2020-05-28	Skoda	1
2020-05-29	Citroen	1
2020-05-29	Honda	1
2020-05-29	Mercedes	1
2020-05-29	Nissan	1
2020-05-30	Dacia	1
2020-05-30	Fiat	1
2020-05-30	Lada	1
2020-05-30	Nissan	1
2020-05-30	Opel	1
2020-05-30	Seat	1
2020-05-31	Alfa Romeo	1
2020-05-31	BMW	1
2020-05-31	Chevrolet	1
2020-05-31	Dacia	1
2020-05-31	Mitsubishi	1
2020-05-31	Peugeot	1
2020-06-01	Chevrolet	1
2020-06-02	Acura	1
2020-06-02	Dacia	1
2020-06-02	Lada	1
2020-06-02	Seat	1
2020-06-03	Acura	1
2020-06-03	Audi	1
2020-06-03	BMW	1
2020-06-03	Mazda	1
2020-06-03	Seat	1
2020-06-04	Acura	1
2020-06-04	Citroen	1
2020-06-04	Kia	1
2020-06-05	Chevrolet	1
2020-06-05	Citroen	1
2020-06-05	Dacia	1
2020-06-05	Honda	1
2020-06-05	Kia	1
2020-06-05	Peugeot	1
2020-06-05	Seat	1
2020-06-06	Audi	1
2020-06-06	Chevrolet	1
2020-06-06	Citroen	1
2020-06-06	Dacia	1
2020-06-06	Honda	1
2020-06-06	Rover	1
2020-06-06	Skoda	1
2020-06-07	Citroen	1
2020-06-07	Dacia	1
2020-06-07	Kia	1
2020-06-07	Peugeot	1
2020-06-07	Skoda	1
2020-06-07	Tofas	1
2020-06-08	Audi	1
2020-06-08	Dacia	1
2020-06-08	Land Rover	1
2020-06-08	Mitsubishi	1
2020-06-09	Audi	1
2020-06-09	Chevrolet	1
2020-06-09	Citroen	1
2020-06-09	Honda	1
2020-06-09	Lada	1
2020-06-09	Skoda	1
2020-06-10	Chevrolet	1
2020-06-10	Citroen	1
2020-06-10	Jeep	1
2020-06-10	Kia	1
2020-06-10	Seat	1
2020-06-11	Audi	1
2020-06-11	BMW	1
2020-06-11	Citroen	1
2020-06-11	Jaguar	1
2020-06-12	Jeep	1
2020-06-12	Lada	1
2020-06-12	Mitsubishi	1
2020-06-12	Porsche	1
2020-06-12	Seat	1
2020-06-13	Audi	1
2020-06-13	Jeep	1
2020-06-13	Seat	1
2020-06-14	Acura	1
2020-06-14	Audi	1
2020-06-14	Land Rover	1
2020-06-14	Mazda	1
2020-06-14	Nissan	1
2020-06-15	Alfa Romeo	1
2020-06-15	Citroen	1
2020-06-15	Honda	1
2020-06-15	Kia	1
2020-06-15	Lada	1
2020-06-15	Mazda	1
2020-06-15	Mercedes	1
2020-06-15	Seat	1
2020-06-16	Acura	1
2020-06-16	Alfa Romeo	1
2020-06-16	Jeep	1
2020-06-16	Tofas	1

Renault is been in advertisements more than 6 times of the next brand. The reasons of this situation can be listed like below:

1. The data may not be collected randomly. (We don’t have any information about the collection of the sample. Maybe, there was a sample selection bias, that the data is sorted with respect to a feature and selected the top 9044 rows of them or this data is collected from only one platform and other platforms are discarded. Also, there are some popular car companies in Turkey market but they are not in the dataset like Toyota, this can be the sign of the sample selection bias. But, to be able to make an analysis, we will assume that this process is done randomly.)
2. In the internet advertisements, Renault is the most appealing brand among others due to strategies of companies of brands (which also we should assume that there is no possibility like that)
3. Prices of Renault’s car can be affordable respect to other cars of brands because of high price increase in taxes.
4. Renault is the most popular Brand in online Turkey market because the company produces cars for diverse segments of consumers which helps to reach higher number of sales.

carmarket_brand = carmarket %>%
  group_by(Brand) %>%
  summarise(count = n(), min_price = min(Price), max_price = max(Price), avg_price = mean(Price), median_price = median(Price))

carmarket_brand  %>%
  arrange(desc(count)) %>%
  select(Brand, count)%>%
  kable(col.names = c("Brand", "Count")) %>%
  kable_styling("striped", full_width = T) %>%
  scroll_box(width = "100%", height = "400px")

Brand	Count
Renault	2079
Fiat	653
Opel	643
Hyundai	638
Ford	601
BMW	598
Mercedes	542
Audi	425
Peugeot	421
Dacia	323
Honda	261
Skoda	259
Nissan	229
Citroen	199
Land Rover	171
Kia	164
Seat	150
Chevrolet	125
Jeep	83
Tofas	45
Acura	41
Porsche	39
Jaguar	28
Mitsubishi	28
Mazda	22
Volkswagen	18
Lada	16
Alfa Romeo	14
Chrysler	11
Infiniti	3
Rover	3
Geely	2

The number of advertisement belong to different car brands are given above table.

carmarket_brand %>%
  arrange(avg_price)%>%
  select(Brand, avg_price, median_price, min_price, max_price) %>%
  kable(col.names = c("Brand", "Average Price", "Median Price", "Minimum Price", "Maximum Price"))%>% 
  kable_styling("striped", full_width = T) %>%
  scroll_box(width = "100%", height = "400px")

Brand	Average Price	Median Price	Minimum Price	Maximum Price
Tofas	17234.44	16500.0	10500	28500
Lada	22143.75	13125.0	5500	119900
Rover	24666.67	24000.0	21000	29000
Geely	43250.00	43250.0	35000	51500
Mazda	66086.36	38675.0	11900	179750
Fiat	68145.62	65900.0	12345	169500
Mitsubishi	73462.50	69850.0	23000	187950
Citroen	79485.17	72500.0	13750	281850
Dacia	81834.67	72500.0	21000	170000
Renault	85985.26	81650.0	9500	300000
Opel	88746.34	82000.0	16000	300000
Chevrolet	89637.99	70000.0	30500	900000
Hyundai	90242.06	84900.0	14750	318000
Ford	94712.37	90000.0	6250	339000
Chrysler	95400.00	96500.0	35000	169500
Honda	106317.52	93000.0	10000	289000
Peugeot	107926.51	82900.0	12750	334000
Alfa Romeo	118064.29	106950.0	32000	343250
Kia	118527.12	122250.0	16500	240000
Seat	121214.98	119000.0	24900	345000
Volkswagen	123500.00	123500.0	123500	123500
Acura	124303.66	148750.0	19750	345000
Skoda	128163.50	131500.0	9900	320000
Nissan	131742.66	127000.0	11111	315000
Infiniti	136966.67	122500.0	78500	209900
Jeep	201382.41	133000.0	42250	457500
Audi	243041.13	191950.0	15000	2550000
Mercedes	290055.56	228874.5	13500	2720000
BMW	294612.95	249000.0	23000	1715000
Jaguar	538166.07	634500.0	54500	634500
Land Rover	604241.52	444500.0	43500	2425000
Porsche	968731.79	779990.0	159500	2375000

When we examine these two tables, we see that there are cheaper cars in the dataset. So, the third option is not true. So, we can assume that Renault is one of the online market leaders in the Turkey market. For the diversity of segments, we need too plot the prices of Renault.

carmarket[Brand == 'Renault', .(Price)] %>%
  ggplot(., aes(y = Price)) +
  geom_boxplot() +
  theme_minimal() +
  expand_limits(x=c(-0.5,0.5)) +
  labs(title = "Price Diversity of Renault",
       subtitle = st,
       y = "Price") +
  scale_y_continuous(labels = comma)

We can see from the plot that Renault has cars for every segment in Turkey market. This could make them to be very popular in the market.

2.2 The Most/Least Popular Brands

In this data set is obtained from the online sales web-page. To find most popular car brand, i.e., the car brands which have more advertisement, we calculate percentage of the cars according to the their Brands. Then,we obtain column chart.

carmarket %>%
  count(Brand, sort=TRUE) %>%
  mutate(percentage = 100 * n / sum(n)) %>%
  head(15) %>%
  ggplot(.,aes(x=percentage, y=reorder(Brand,percentage), fill = percentage)) +
  geom_col() +
  scale_fill_gradient("percentage", low="seagreen2", high="seagreen4") +
  geom_text(aes(label = paste(format(percentage,digits=3), "%")), size=4, position = position_stack(vjust = 0.5)) +
  theme_minimal() +
  theme(legend.position = "none", plot.title = element_text(vjust = 0.5)) +
  labs(x = "Percentages",
       y = "Car Brands",
       title = "Top 15 Popular Brands",
       subtitle = st)

The result shows that, the most popular brands according to the number of advertisement in the online car market is Renault. It consists 23% of the advisement and Fiat is following this brand. This result also shows that Renault is one of the most preferable car brand in Turkey. For this reason, it can be sold more than the other Brands.

carmarket %>%
  count(Brand, sort=TRUE) %>%
  mutate(percentage = 100 * n / sum(n)) %>%
  tail(15) %>%
  ggplot(.,aes(x=percentage, y=reorder(Brand,percentage), fill=percentage)) +
  geom_col() +
  scale_fill_gradient("percentage", low="peachpuff2", high="peachpuff4") +
  geom_text(aes(label = paste(format(percentage,digits=1), "%")), size=4, position = position_stack(vjust = 0.5)) +
  theme_minimal() +
  theme(legend.position = "none", plot.title = element_text(vjust = 0.5)) +
  labs(x = "Percentages",
       y = "Car Brands",
       title = "Less Popular Brands",
       subtitle = st)

Like the the most popular brands, we can also sort the least popular brands in the online car market. The percentages of the Brands are calculated by using number of cars according to the Brands. The brands in this result indicates that the most expensive car brands like Maserati, Jaguar have smaller percentage because they are expensive car brands, for this reason there are few owner of this car, less advertisements.

Note that, although the calculation is same with the the most popular car brands, because of the graph scale, the Tofas can be seen as higher percentage. However, when we examine the percentage value, we can get real results.

2.2 Price Analysis of Car Brands

In the data set, the each car its own price, and these various price can be classified by using price intervals. By using quantile, we divide price variable into five level and then, we sort from the Very Low to the Very High. And we create another variable for this data set, which is called Price Group. After that, by using this classification, cars are examined with their price groups.

quant = quantile(carmarket$Price, seq(0, 1, 0.2))

carmarket_price_group <- carmarket %>%
  mutate(price_group = case_when(
    Price < quant[2] ~ "Very Low",
    Price < quant[3] ~ "Low",
    Price < quant[4] ~ "Medium",
    Price < quant[5] ~ "High",
    TRUE ~ "Very High"
  )) %>%
  mutate(price_group = factor(price_group, levels = c("Very Low", "Low", "Medium", "High", "Very High")))

carmarket_price_group %>%
  group_by(Brand, price_group) %>%
  summarize(counter = n()) %>%
  mutate(percentage = 100 * counter / sum(counter)) %>%
  ggplot(., aes(x = '', y = counter, fill = price_group)) + 
  geom_bar(width = 1, stat = "identity", position = "fill") +
  coord_polar("y") +
  theme_void() +
  theme(plot.title = element_text(vjust = 0.5)) +
  facet_wrap(~Brand) +
  labs(title = "Price Group Analyses of Car Brand",
       subtitle = st,
       fill = "Price Group")

By using Price Group classification, we can obtain spread of the price in each car brand. Geely, and Tofas, for example, have only very low price, whereas the Maserati’s price is in the very high class. Moreover, Jaguar and Porsche are almost in the very high class. The others have different price interval.

To give more understandable analysis for the price of the brands, the following table can be used. This table illustrates the minimum, maximum and average prices of the cars according to the brands.

carmarket %>%
  group_by(Brand) %>%
  summarize(MinPrice = min(Price),
            Average_Price = round(mean(Price)),
            MaxPrice = max(Price)) %>%
  select(Brand,MinPrice, Average_Price,MaxPrice) %>%
  arrange(desc(Average_Price)) %>%
  head(10) %>%
  kable(col.names = c("Brand",  "Minimum Price", "Average Price", "Maximum Price")) %>%
  kable_styling("striped", full_width = T) %>%
  scroll_box(width = "100%", height = "400px")

Brand	Minimum Price	Average Price	Maximum Price
Porsche	159500	968732	2375000
Land Rover	43500	604242	2425000
Jaguar	54500	538166	634500
BMW	23000	294613	1715000
Mercedes	13500	290056	2720000
Audi	15000	243041	2550000
Jeep	42250	201382	457500
Infiniti	78500	136967	209900
Nissan	11111	131743	315000
Skoda	9900	128163	320000

The Brands are sorted according to the average price. In the first place, there is Porsche. This results also support the price group analysis. Moreover, to provide more understandable interpretation for the Brand, we present the bar chart of the Brands according to the their average price.

carmarket %>%
  group_by(Brand) %>%
  summarise(Average_Price = mean(Price)) %>%
  arrange(desc(Average_Price))%>%
  
  ggplot(., aes(y=reorder(Brand,Average_Price), x = Average_Price, fill = Brand)) +
    geom_col() +
    theme_minimal() +
    labs(title = "Order of the Car Brand According to \nTheir Average Price",
         subtitle = st,
         x = "Average Price",
         y = "Brands")

2.3 Price Analysis By Body Type

In the previous section, we analyze the car brands and their features like Price. In this section we present the body type of the cars. One important factor that impacts this decision is the type of car body which refers to the shape and the model.

carmarket %>%
  group_by(Body_Type) %>%
  summarize(Average_Price = mean(Price)) %>%
  
  ggplot(.,aes(x=reorder(Body_Type, -Average_Price), y = Average_Price, color= Body_Type)) +
    geom_point(size=7) +
    geom_segment(aes(x=Body_Type,
                     xend=Body_Type,
                     y=0,
                     yend=Average_Price))+
    theme_minimal() +
    scale_y_continuous(labels = function(x) format(x, scientific = FALSE)) +
    theme(axis.text.x = element_text(angle = 90), legend.position = "none")+
        labs(title = "Average Price According to the Body Type",
         subtitle = st, 
         x = "Body Type",
         y = "Average Price",
         Fill = "Body Type")

In the first place, there is sport/coupe body type because of the high average price of these cars. The reason behind this idea is that, the most of luxury cars with high price are sport cars such as Porsche.

2.4 Price Analysis By Fuel Type

The fuel of the cars can be varied. There are for different fuel type: (i) diesel, (ii) electricity, (iii) gasoline, and (iv) hybrid. While diesel is produced by the distillation of crude oil, Gasoline is obtained from crude oil and other petroleum liquids. Hybrids, on the other hand, is the combination of the gasoline and electrical.

carmarket %>%
  group_by(Fuel_Type) %>%
  summarise(count = n()) %>%
  mutate(percentage = 100*round(count/sum(count),3)) %>%
  
  ggplot(., aes(x = '', y = count, fill = Fuel_Type)) + 
  geom_bar(width = 1, stat = "identity") +
  coord_polar("y") +
  theme_void()+
  theme(plot.title = element_text(vjust = 0.5)) +
  geom_text(aes(label = paste(format(percentage,digits=2), "%")), size=4, position = position_stack(vjust = 0.5)) +
  labs(title = "Percentages of Fuel Types",
       subtitle = st,
       fill = "Fuel Types")

The pie chart shows that, the most of the cars in this dataset use Diesel, whereas the Hybrid usage is very low. These different fuel types can affect the price of the cars. To examine this assumption, we calculate minimum, maximum, and average price that are given below according to the fuel type.

carmarket %>%
  group_by(Fuel_Type) %>%
  summarize(count = n(),
            Min_Price = min(Price),
            Average_Price = mean(Price),
            Max_Price = max(Price)) %>%
  mutate(percentage = 100*round(count/sum(count),3))%>%
  arrange(desc(Average_Price)) %>%
  select(Fuel_Type, count,  percentage, Min_Price, Average_Price, Max_Price ) %>%
  kable(col.names = c("Fuel Type", "Number of Ad.", "Percentage", "Minimum Price", "Average Price", "Maximum Price"))  %>%
  kable_minimal(full_width = F)

Fuel Type	Number of Ad.	Percentage	Minimum Price	Average Price	Maximum Price
Hybrid	27	0.3	126900	892905.56	2375000
Electricity	1348	15.3	10000	191744.10	2550000
Diesel	5812	65.8	11000	149461.04	2720000
Gasoline	1647	18.6	5500	58290.84	293000

When we consider the table, the order of the fuel types according to the average price from the lowest to highest is Gasoline, Diesel, Electricity, and Hybrid. Moreover, the spread of price according to the fuel types is illustrated as box plot below.

carmarket %>%
  group_by(Fuel_Type) %>%

  ggplot(.,aes(x=Fuel_Type, y = Price, fill= Fuel_Type)) +
    geom_boxplot() +
    theme_minimal() +
    scale_y_continuous(labels = function(x) format(x, scientific = FALSE)) +
    labs(title = "Prices According to the Fuel Type",
         subtitle = st, 
         x = "Fuel Type",
         y = "Price",
         fill = "Fuel Type")

• There are many outliers in Diesel and Electricity Fuel Type.
• Gasoline is the least expensive fuel type, on the other hand, hybrid is the most expensive one.
• The median of the Fuel Types, which are Diesel, Electricity, and Gasoline, almost in the middle of the quantiles. However, the median Hybrid is very close to the first quantile.
• Moreover, since the price interval of the Hybrid is wider, there is no outlier data.

2.5 Price Analysis By Gear

Like the fuel type, there is another feature, Gear Type, which affects the selection of cars. For this, we search the relationship between price and gear type.

carmarket %>%
  ggplot(., aes(x = Gear, y = Price, color = Gear)) +
  geom_jitter() +
  theme_minimal() +
  scale_y_continuous(labels = function(x) format(x, scientific = FALSE)) +
    labs(title = " Price According to the Gear Type",
         x = "Gear Type",
         y = "Price",
         color = "Gear Type")

The half of the cars in this data set belongs to the Manual Gear Type. However, price distribution in manual gear type concentrates in a narrow area. Semi automatic and automatic, on the other hand, have a wider area with possible outliers.

carmarket %>%
  group_by(Gear) %>%
  summarize(count=n(),Min_Price=min(Price),Average_Price = mean(Price),Max_Price=max(Price) ) %>%
  mutate(percentage = 100*round(count/sum(count),3))%>%
  arrange(desc(percentage)) %>%
  select(Gear, count,  percentage, Min_Price, Average_Price, Max_Price ) %>%
  kable(col.names = c("Gear","Number of Ad","Percentage",  "Minimum Price", "Average Price", "Maximum Price")) %>%
  kable_minimal(full_width = F)

Gear	Number of Ad	Percentage	Minimum Price	Average Price	Maximum Price
Manual	4462	50.5	5500	68453.65	300000
Automatic	2325	26.3	10000	240026.97	2720000
Semi Automatic	2047	23.2	36000	187469.05	1550000

We give gear type branching according to the car brands below.

carmarket %>%
  group_by(Brand, Gear) %>%
  summarize(gear_type_count = n())  %>%
  mutate(gear_type_percentage = 100*round(gear_type_count / sum(gear_type_count), digits = 3)) %>%
  select(Brand, Gear, gear_type_percentage) %>%
  pivot_wider(id_cols = Brand, names_from = Gear, values_from = gear_type_percentage) %>%
  kable(col.names = c("Brand", "Automatic (%)", "Manual (%)", "Semi Automatic (%)"))%>%
  kable_styling("striped", full_width = T) %>%
  scroll_box(width = "100%", height = "400px")

Brand	Automatic (%)	Manual (%)	Semi Automatic (%)
Acura	48.8	46.3	4.9
Alfa Romeo	35.7	42.9	21.4
Audi	33.2	3.5	63.3
BMW	50.8	7.5	41.6
Chevrolet	39.2	52.0	8.8
Chrysler	100.0	NA	NA
Citroen	13.6	71.4	15.1
Dacia	0.9	96.9	2.2
Fiat	4.3	84.1	11.6
Ford	13.3	69.2	17.5
Geely	NA	100.0	NA
Honda	52.5	33.0	14.6
Hyundai	32.1	53.3	14.6
Infiniti	66.7	NA	33.3
Jaguar	92.9	3.6	3.6
Jeep	92.8	7.2	NA
Kia	53.0	31.1	15.9
Lada	NA	100.0	NA
Land Rover	74.3	1.2	24.6
Mazda	50.0	50.0	NA
Mercedes	57.9	12.7	29.3
Mitsubishi	14.3	78.6	7.1
Nissan	38.4	41.5	20.1
Opel	34.8	57.9	7.3
Peugeot	31.1	57.0	11.9
Porsche	51.3	NA	48.7
Renault	6.5	67.2	26.3
Rover	33.3	66.7	NA
Seat	10.0	37.3	52.7
Skoda	13.1	30.5	56.4
Tofas	NA	100.0	NA
Volkswagen	100.0	NA	NA

2.6 Price Analysis with Gear and Fuel Type

Up until now, we examine relationship between a variable with price. In this subsection, we address relationship between price and more than one variable, i.e., Gear and Fuel Type.

carmarket %>%
  group_by(Fuel_Type, Gear) %>%
  summarise(Average_Price = mean(Price)) %>%
  
  ggplot(.,aes(x=reorder(Fuel_Type, -Average_Price), y = Average_Price, fill= Fuel_Type)) +
    geom_col() +
    facet_wrap(~Gear) +
    theme_minimal() +
    theme(axis.text.x = element_text(angle = 90), legend.position = "none") +
    scale_y_continuous(labels = function(x) format(x, scientific = FALSE)) +
    labs(title = "Average Price According to the Fuel Type for Different Gear Type",
         subtitle = st, 
         x = "Fuel Type",
         y = "Average Price",
         Fill = "Fuel Type")

According to the results,

• The most expensive cars are located in Semi-Automatic and Automatic Gear Type with Hybrid Fuel Type.
  
• While the Hybrid Fuel Type is more recent technology, there are no cars in Manual Gear Type.
• The least expensive cars are located in Gasoline Fuel type in all Gear Types.

2.7 Price Analysis By CCM

The CCM is also important to predict car price, for this reason before to create linear regression model, we want to analyze relationship between Price and CCM.

carmarket%>% 
  group_by(CCM)%>% 
  ggplot(., aes(x=CCM, y=Price, fill=CCM))+
  geom_boxplot()+
  theme_minimal()+
  scale_y_continuous(labels = function(x) format(x, scientific = FALSE)) +
  scale_y_log10()+
  theme(axis.text.x = element_text(angle = 90), legend.position = "none")+
    labs(title = "Average Price According to CCM",
         subtitle = st, 
         x = "CCM",
         y = "Price",
         Fill = "CCM")

2.8 Price Analysis By HP and Seller Status

Horsepower is a unit of power used to measure the forcefulness of a car’s engine.We want to analyze relationship between Price and Horse Power by using Seller Status.

carmarket %>%
  group_by(Horse_Power, Seller_Status) %>%
  summarize(mean_price = mean(Price)) %>%
  
  ggplot(., aes(y=mean_price, x = Horse_Power, fill = Horse_Power)) +
  geom_col() +
  facet_wrap(~Seller_Status) +
  theme_minimal() +
  theme(legend.position = "none") +
  scale_y_continuous(labels = function(x) format(x, scientific = FALSE)) +
  theme(axis.text.x = element_text(angle = 90), legend.position = "none")+
  labs(title = "Average Price According to the Horse Power",
       subtitle = st, 
       x = "Average Price",
       y = "Horse Power")

carmarket %>%
  group_by(Horse_Power) %>%
  summarize(count=n(),
            Average_Price = mean(Price)) %>%
  mutate(percentage = 100*round(count/sum(count),3))%>%
  arrange(desc(percentage)) %>%
  select(Horse_Power, count,  percentage, Average_Price ) %>%
  kable(col.names = c("Horse Power","Number of Ad","Percentage", "Average Price")) %>%
  kable_minimal(full_width = F)

Horse Power	Number of Ad	Percentage	Average Price
100 HP and below	5440	61.6	97748.33
101-125 HP	1315	14.9	119633.04
126-150 HP	892	10.1	171425.86
151-175 HP	484	5.5	256803.36
176-200 HP	267	3.0	410941.04
276-300 HP	202	2.3	294607.92
201-225 HP	70	0.8	387050.70
226-250 HP	68	0.8	603539.56
251-275 HP	70	0.8	485320.71
301-325 HP	19	0.2	374136.84
326-350 HP	4	0.0	593475.00
376-400 HP	1	0.0	56750.00
451-475 HP	1	0.0	41000.00
601 HP and above	1	0.0	60000.00

• There are no car advertisement with more than 200 Horse Power in Classic and Damaged Seller Status.
• Although there are some car advertisements in 2nd Hand Seller Status, the majority is in 0 Km.
• While the 2nd Hand cars category has almost all Horse Power Types, O Km cars category does not contain Horse Power over 325 HP.

2.9 Seller and Seller Status Relationship

In this online car market data set, there are various sellers and these sellers present different car status such as 0 km, 2nd Hand, Classic, and Damaged. To see how many status takes place in different seller, we create this subsection.

carmarket %>%
  group_by(Seller,Seller_Status) %>%
  summarise(count=n()) %>%
  mutate(percentage = 100*count/sum(count)) %>%

  
  ggplot(., aes(x = '', y = count, fill = Seller_Status)) +
  geom_bar(width = 1, stat = "identity", position = "fill") +
  coord_polar("y") +
  theme_void() +
  theme(plot.title = element_text(vjust = 0.5)) +
  facet_wrap(~Seller) +
  labs(title = "Seller Status Distribution of Sellers",
       subtitle = st,
       fill = "Seller Status")

• Gallery and Owner mostly sell 2nd Hand cars, whereas Authority sell 0 km cars.
• There are no seller status, which are Classic and Damaged in the Authority Seller.

carmarket %>%
  group_by(Seller,Seller_Status) %>%
  summarise(count=n()) %>%
  mutate(percentage = 100*round(count/sum(count), digits = 3))%>%
  select(Seller, Seller_Status, count, percentage)%>%
  kable(col.names = c("Seller", "Seller Status", "Count", "Percentage")) %>%
  kable_minimal(full_width = F)

Seller	Seller Status	Count	Percentage
Authority	0 km	63	94.0
Authority	2nd Hand	4	6.0
Galery	0 km	159	2.2
Galery	2nd Hand	6957	97.3
Galery	Classic	3	0.0
Galery	Damaged	33	0.5
Owner	0 km	12	0.7
Owner	2nd Hand	1590	98.5
Owner	Classic	7	0.4
Owner	Damaged	6	0.4

To provide clear numerical information about Seller and Seller Status, we also give the number and percentage advertisements according to the Seller and Seller Status.

2.10 Gear and Brand Relationship

In this part, we give the relationship between Gear Type and Brand.

carmarket %>%
  group_by(Brand, Gear) %>%
  summarize(gear_type_count = n())  %>%
  mutate(gear_type_percentage = gear_type_count / sum(gear_type_count)) %>%
  #head(108) %>%
  ggplot(., aes(x = Brand, y = gear_type_count, fill = Gear)) +
  geom_bar(position = "fill",stat = "identity") +
  theme_minimal() +
  scale_y_continuous(labels = scales::percent_format()) +
  #geom_text(aes(label = format(gear_type_percentage, digits=3)), size=4, position = position_dodge(0.9)) +
  theme(axis.text.x = element_text(angle = 90), legend.position = "right") +
  labs(title = "Gear Type Comparison of Car Brands",
       subtitle = st,
       x = "Brand",
       y = "Percentage of Gear Type", 
       fill = "Gear Type")

The results show that,

• The all cars of the Chyrsler and Volkswagen have only automatic gear type, in the 2020 online car market.
• Geely, Lada, and Tofas have only manual gear type.
• The other car brands have various gear type in their cars.

2.11 The Most Popular Car Colors

carmarket %>%
  group_by(Color) %>%
  summarize(count=n(),Min_Price=min(Price),Average_Price = mean(Price),Max_Price=max(Price) ) %>%
  mutate(percentage = 100*round(count/sum(count),3))%>%
  arrange(desc(percentage)) %>%
  head(10) %>%
  ggplot(.,aes(x=percentage, y=reorder(Color,percentage), fill=percentage)) +
  geom_col() +
  scale_fill_gradient("percentage", low="thistle1", high="thistle4") +
  geom_text(aes(label = paste(format(percentage,digits=1), "%")), size=4, position = position_stack(vjust = 0.5)) +
  theme_minimal() +
  theme(legend.position = "none", plot.title = element_text(vjust = 0.5)) +
  labs(x = "Percentages",
       y = "Colors",
       title = "The Most Popular Colors",
       subtitle = st)

carmarket %>%
  group_by(Color) %>%
  summarize(count=n(),Min_Price=min(Price),Average_Price = mean(Price),Max_Price=max(Price) ) %>%
  mutate(percentage = 100*round(count/sum(count),3))%>%
  arrange(desc(percentage)) %>%
  #head(10) %>%
  select(Color, count,  percentage, Min_Price, Average_Price, Max_Price ) %>%
  kable(col.names = c("Color","Number of Ad","Percentage",  "Minimum Price", "Average Price", "Maximum Price")) %>%
  kable_styling("striped", full_width = T) %>%
  scroll_box(width = "100%", height = "400px")

Color	Number of Ad	Percentage	Minimum Price	Average Price	Maximum Price
White	4224	47.8	9500	132096.68	1980000
Black	1159	13.1	10000	259959.36	2720000
Gray	1004	11.4	11500	120185.29	1550000
Silver Gray	640	7.2	10000	77366.37	553000
Smoked	420	4.8	10000	120774.66	900000
Red	379	4.3	5500	120049.33	899950
Blue	320	3.6	10000	111402.49	1815000
Dark Blue	163	1.8	14650	187614.11	634500
Burgundy	119	1.3	6250	70623.11	485000
Brown	88	1.0	10000	171926.68	950000
Green	80	0.9	9750	64659.99	334000
Beige	61	0.7	12500	139031.97	1190000
Other	40	0.5	29000	112575.00	236750
Orange	32	0.4	29000	187115.62	399500
Yellow	23	0.3	22000	93863.04	387950
Champagne	20	0.2	15000	76687.50	258000
Silver	20	0.2	19500	69549.00	212000
Honey	10	0.1	16000	66575.00	161000
Purple	5	0.1	16500	61530.00	121500
Sand Color	6	0.1	15500	83791.67	267000
Turquoise	10	0.1	17250	64624.90	175000
Amaranth	1	0.0	35500	35500.00	35500
Cream	2	0.0	13500	14950.00	16400
Gold	2	0.0	31300	38150.00	45000
Linden	2	0.0	10000	83500.00	157000
Magenta	1	0.0	20000	20000.00	20000
Olive Gray	2	0.0	15500	53750.00	92000
Pink	1	0.0	28000	28000.00	28000

According to the result, the most popular color in car is White. The second one is Black. These results are as expected.

3. Conclusion

In this Exploratory Data Analysis we investigate the 2020 online car market data in Turkey. The main purpose of this study is to find the relationship between features and Price of the cars. Each car has different properties, for example body type, gear type, color, etc. These properties affect both the Price and the demand of car market in Turkey. For this reason, before conducting models for forecasting car price, first, variables and their sub-groups are analyzed. Then, graphical visualizations are introduced to present more understandable information about dataset. In the next section, some models to forecast price of cars will be addressed.

References

Kaggle Notebook
Color Cheatsheet
Car Feature Explanations
Kable Styling
ggplot visualizations
Lecture Notes