Car Price Prediction for WQD7004 Programming for Data Science
Group 11
2024-06-09
1 Group Members
| Student ID | Name | Role |
|---|---|---|
| 17061830 | Nadzirah Raihan | Discussions & Conclusion |
| 22104060 | Zhang Wei | Introduction & Project Objective |
| 23059592 | Boaz Chung Yi Heng | Data Understanding |
| 23069683 | Diva Alifta Chandra | Data Analysis - Regressor |
| 23083416 | Yong Ting Kang | Data Analysis - Classifier |
| 23072236 | Sim Jin Xiang | Data Cleaning |
2 Introduction
When exploring strategic approaches in the automotive industry, utilizing machine learning models to understand and predict car pricing can significantly enhance decision-making and optimize business operations.
This project presents a comprehensive analysis of employing both classification and regression models to dissect the intricacies of car pricing based on a rich dataset that includes a variety of car attributes such as make, fuel type, body style, and more importantly, price.
The dataset provides a robust platform for applying these models, containing detailed features that describe each car’s specifications. By analyzing these attributes, we aim to classify cars into distinct price segments—high-end and mid-low-end—using classification algorithms. This segmentation will allow for targeted marketing strategies and precise customer segmentation, crucial for competitive advantage in a saturated market.
Furthermore, we explore the use of regression models to predict the exact price of cars based on their features. This predictive modeling will help in setting accurate prices, essential for maintaining profitability and competitiveness. By predicting prices, we can also gain insights into market trends and the factors influencing car values, which is invaluable for strategic planning and risk management.
This project aims to not only demonstrate the application of these models to real-world data but also to discuss the implications of the findings in terms of business strategies and operations optimization in the automotive sector. By leveraging machine learning, automotive businesses can achieve a greater understanding of their product pricing and market dynamics, thereby enhancing their marketing, sales, and development strategies.
2.1 Objectives & Outputs
2.1.1 Using Classification Models to Segment Car Prices
2.1.1.1 Precision Marketing:
By identifying cars in different price segments, companies can more effectively target their market and design customized marketing strategies for different customer groups. For example, high-end cars might focus more on promoting brand image and unique features, while mid-low-end cars might emphasize cost-effectiveness and practical features.
2.1.1.2 Product Positioning and Pricing Strategy:
By categorizing cars into high-end and mid-low-end segments, companies can adjust their product features and pricing strategies to maximize market coverage and profits.
2.1.1.3 Promotional and Sales Strategies:
Classification models enable sales teams to better understand customer purchasing power and preferences, allowing for effective sales and promotional strategies.
2.1.1.4 Consumer Insights:
Analyzing the behaviors and preferences of car buyers across different price segments can provide important insights into consumer demand. This helps businesses better understand market dynamics and design products that meet consumer expectations.
2.1.2 Using Regression Models to Predict Car Prices
2.1.2.1 Accurate Pricing:
Regression models can predict the specific price of cars, providing businesses with a scientific basis for pricing, helping to set more precise prices and avoiding profit losses or reduced market competitiveness due to overpricing or underpricing.
2.1.2.2 Market Trend Analysis:
By analyzing how various factors affect car prices, regression models can reveal trends and causes of price changes, providing strategic decision support for a company’s long-term development.
2.1.2.3 Risk Management:
In financial services such as car loans and insurance, accurate price predictions help assess asset values and associated risks, thereby designing reasonable loan amounts and insurance rates.
2.1.2.4 Supply Chain Optimization:
Predicting the price and market demand for different types of cars enables companies to manage their production plans and inventory more effectively, reducing costs and improving operational efficiency.
3 Loading Necessary Packages
# Load necessary packages
suppressPackageStartupMessages({
library(dplyr)
library(VIM)
library(ggplot2)
library(zoo)
library(gridExtra)
})4 Data Understanding
4.1 Loading Data
This project uses the dataset titled “Automobile” from 1985 Ward’s Automotive Yearbook donated on 18-May-1987.
# Use the existing dataset
data <- read.csv('imports-85.data', sep=',')
# Rename columns if necessary
colnames(data) <- c('symboling', 'normalized.losses','make', 'fuel.type','aspiration',
'num.of.doors','body.style','drive.wheels','engine.location',
'wheel.base','length','width','height','curb.weight','engine.type',
'num.of.cylinders','engine.size','fuel.system','bore','stroke',
'compression.ratio','horsepower','peak.rpm','city.mpg','highway.mpg',
'price')4.2 Initial EDA
This dataset is structured in a tabular format and is made up of 26 columns and 205 rows (or observations). It consists of three types of information: the vehicle specifications, its assigned insurance risk rating, and its normalized losses. The variable names, data types and summary statistics of each variable are described below.
# Dataset dimensions
dim(data)## [1] 204 26# Inspecting the data
head(data)## symboling normalized.losses make fuel.type aspiration num.of.doors
## 1 3 ? alfa-romero gas std two
## 2 1 ? alfa-romero gas std two
## 3 2 164 audi gas std four
## 4 2 164 audi gas std four
## 5 2 ? audi gas std two
## 6 1 158 audi gas std four
## body.style drive.wheels engine.location wheel.base length width height
## 1 convertible rwd front 88.6 168.8 64.1 48.8
## 2 hatchback rwd front 94.5 171.2 65.5 52.4
## 3 sedan fwd front 99.8 176.6 66.2 54.3
## 4 sedan 4wd front 99.4 176.6 66.4 54.3
## 5 sedan fwd front 99.8 177.3 66.3 53.1
## 6 sedan fwd front 105.8 192.7 71.4 55.7
## curb.weight engine.type num.of.cylinders engine.size fuel.system bore stroke
## 1 2548 dohc four 130 mpfi 3.47 2.68
## 2 2823 ohcv six 152 mpfi 2.68 3.47
## 3 2337 ohc four 109 mpfi 3.19 3.40
## 4 2824 ohc five 136 mpfi 3.19 3.40
## 5 2507 ohc five 136 mpfi 3.19 3.40
## 6 2844 ohc five 136 mpfi 3.19 3.40
## compression.ratio horsepower peak.rpm city.mpg highway.mpg price
## 1 9.0 111 5000 21 27 16500
## 2 9.0 154 5000 19 26 16500
## 3 10.0 102 5500 24 30 13950
## 4 8.0 115 5500 18 22 17450
## 5 8.5 110 5500 19 25 15250
## 6 8.5 110 5500 19 25 17710# Variable names and data types
str(data)## 'data.frame': 204 obs. of 26 variables:
## $ symboling : int 3 1 2 2 2 1 1 1 0 2 ...
## $ normalized.losses: chr "?" "?" "164" "164" ...
## $ make : chr "alfa-romero" "alfa-romero" "audi" "audi" ...
## $ fuel.type : chr "gas" "gas" "gas" "gas" ...
## $ aspiration : chr "std" "std" "std" "std" ...
## $ num.of.doors : chr "two" "two" "four" "four" ...
## $ body.style : chr "convertible" "hatchback" "sedan" "sedan" ...
## $ drive.wheels : chr "rwd" "rwd" "fwd" "4wd" ...
## $ engine.location : chr "front" "front" "front" "front" ...
## $ wheel.base : num 88.6 94.5 99.8 99.4 99.8 ...
## $ length : num 169 171 177 177 177 ...
## $ width : num 64.1 65.5 66.2 66.4 66.3 71.4 71.4 71.4 67.9 64.8 ...
## $ height : num 48.8 52.4 54.3 54.3 53.1 55.7 55.7 55.9 52 54.3 ...
## $ curb.weight : int 2548 2823 2337 2824 2507 2844 2954 3086 3053 2395 ...
## $ engine.type : chr "dohc" "ohcv" "ohc" "ohc" ...
## $ num.of.cylinders : chr "four" "six" "four" "five" ...
## $ engine.size : int 130 152 109 136 136 136 136 131 131 108 ...
## $ fuel.system : chr "mpfi" "mpfi" "mpfi" "mpfi" ...
## $ bore : chr "3.47" "2.68" "3.19" "3.19" ...
## $ stroke : chr "2.68" "3.47" "3.40" "3.40" ...
## $ compression.ratio: num 9 9 10 8 8.5 8.5 8.5 8.3 7 8.8 ...
## $ horsepower : chr "111" "154" "102" "115" ...
## $ peak.rpm : chr "5000" "5000" "5500" "5500" ...
## $ city.mpg : int 21 19 24 18 19 19 19 17 16 23 ...
## $ highway.mpg : int 27 26 30 22 25 25 25 20 22 29 ...
## $ price : chr "16500" "16500" "13950" "17450" ...# Summary statistics
summary(data)## symboling normalized.losses make fuel.type
## Min. :-2.0000 Length:204 Length:204 Length:204
## 1st Qu.: 0.0000 Class :character Class :character Class :character
## Median : 1.0000 Mode :character Mode :character Mode :character
## Mean : 0.8235
## 3rd Qu.: 2.0000
## Max. : 3.0000
## aspiration num.of.doors body.style drive.wheels
## Length:204 Length:204 Length:204 Length:204
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
## engine.location wheel.base length width
## Length:204 Min. : 86.60 Min. :141.1 Min. :60.30
## Class :character 1st Qu.: 94.50 1st Qu.:166.3 1st Qu.:64.08
## Mode :character Median : 97.00 Median :173.2 Median :65.50
## Mean : 98.81 Mean :174.1 Mean :65.92
## 3rd Qu.:102.40 3rd Qu.:183.2 3rd Qu.:66.90
## Max. :120.90 Max. :208.1 Max. :72.30
## height curb.weight engine.type num.of.cylinders
## Min. :47.80 Min. :1488 Length:204 Length:204
## 1st Qu.:52.00 1st Qu.:2145 Class :character Class :character
## Median :54.10 Median :2414 Mode :character Mode :character
## Mean :53.75 Mean :2556
## 3rd Qu.:55.50 3rd Qu.:2939
## Max. :59.80 Max. :4066
## engine.size fuel.system bore stroke
## Min. : 61.0 Length:204 Length:204 Length:204
## 1st Qu.: 97.0 Class :character Class :character Class :character
## Median :119.5 Mode :character Mode :character Mode :character
## Mean :126.9
## 3rd Qu.:142.0
## Max. :326.0
## compression.ratio horsepower peak.rpm city.mpg
## Min. : 7.000 Length:204 Length:204 Min. :13.00
## 1st Qu.: 8.575 Class :character Class :character 1st Qu.:19.00
## Median : 9.000 Mode :character Mode :character Median :24.00
## Mean :10.148 Mean :25.24
## 3rd Qu.: 9.400 3rd Qu.:30.00
## Max. :23.000 Max. :49.00
## highway.mpg price
## Min. :16.00 Length:204
## 1st Qu.:25.00 Class :character
## Median :30.00 Mode :character
## Mean :30.77
## 3rd Qu.:34.50
## Max. :54.005 Data Cleaning
#We can see that many attributes are not of the correct datatype.
# Replace '?' with NA
data[data == '?'] <- NA
# Identify columns with at least one NA value
columns_with_na <- colnames(data)[colSums(is.na(data)) > 0]
data_filtered <- data[, columns_with_na]
# Visualize missing data
aggr(data_filtered, numbers = TRUE, sortVars = TRUE, labels = colnames(data_filtered), col = c('red', 'blue'), cex.axis = 0.55)
##
## Variables sorted by number of missings:
## Variable Count
## normalized.losses 0.196078431
## bore 0.019607843
## stroke 0.019607843
## price 0.019607843
## num.of.doors 0.009803922
## horsepower 0.009803922
## peak.rpm 0.0098039225.1 Findings
# Convert 'normalized.losses' to numeric (if it's not already)
data$normalized.losses <- as.numeric(as.character(data$normalized.losses))
# Create a histogram after filtering out NA values
ggplot(data[!is.na(data$normalized.losses), ], aes(x = normalized.losses)) +
geom_histogram(binwidth = 5,color="blue",fill="lightgreen") +
xlab("Normalized Losses") +
ylab("Count")
# Calculate summary statistics by symboling
summary_stats <- aggregate(normalized.losses ~ symboling, data, function(x) {
c(count = length(na.omit(x)),
mean = mean(x, na.rm = TRUE),
min = min(x, na.rm = TRUE),
Q1 = quantile(x, probs = 0.25, na.rm = TRUE),
median = median(x, na.rm = TRUE),
Q3 = quantile(x, probs = 0.75, na.rm = TRUE),
max = max(x, na.rm = TRUE))
})
print(summary_stats)## symboling normalized.losses.count normalized.losses.mean
## 1 -2 3.0000 103.0000
## 2 -1 20.0000 85.6000
## 3 0 48.0000 113.1667
## 4 1 47.0000 128.5745
## 5 2 29.0000 125.6897
## 6 3 17.0000 168.6471
## normalized.losses.min normalized.losses.Q1.25% normalized.losses.median
## 1 103.0000 103.0000 103.0000
## 2 65.0000 71.7500 91.5000
## 3 77.0000 91.0000 102.0000
## 4 74.0000 105.5000 125.0000
## 5 83.0000 94.0000 134.0000
## 6 142.0000 150.0000 150.0000
## normalized.losses.Q3.75% normalized.losses.max
## 1 103.0000 103.0000
## 2 95.0000 137.0000
## 3 120.5000 192.0000
## 4 148.0000 231.0000
## 5 137.0000 192.0000
## 6 194.0000 256.00005.2 Drop NA values from specific columns
data2 <- data[complete.cases(data[, c("price", "bore", "stroke", "peak.rpm", "horsepower", "num.of.doors")]), ]
data2## symboling normalized.losses make fuel.type aspiration num.of.doors
## 1 3 NA alfa-romero gas std two
## 2 1 NA alfa-romero gas std two
## 3 2 164 audi gas std four
## 4 2 164 audi gas std four
## 5 2 NA audi gas std two
## 6 1 158 audi gas std four
## 7 1 NA audi gas std four
## 8 1 158 audi gas turbo four
## 10 2 192 bmw gas std two
## 11 0 192 bmw gas std four
## 12 0 188 bmw gas std two
## 13 0 188 bmw gas std four
## 14 1 NA bmw gas std four
## 15 0 NA bmw gas std four
## 16 0 NA bmw gas std two
## 17 0 NA bmw gas std four
## 18 2 121 chevrolet gas std two
## 19 1 98 chevrolet gas std two
## 20 0 81 chevrolet gas std four
## 21 1 118 dodge gas std two
## 22 1 118 dodge gas std two
## 23 1 118 dodge gas turbo two
## 24 1 148 dodge gas std four
## 25 1 148 dodge gas std four
## 26 1 148 dodge gas std four
## 28 -1 110 dodge gas std four
## 29 3 145 dodge gas turbo two
## 30 2 137 honda gas std two
## 31 2 137 honda gas std two
## 32 1 101 honda gas std two
## 33 1 101 honda gas std two
## 34 1 101 honda gas std two
## 35 0 110 honda gas std four
## 36 0 78 honda gas std four
## 37 0 106 honda gas std two
## 38 0 106 honda gas std two
## 39 0 85 honda gas std four
## 40 0 85 honda gas std four
## 41 0 85 honda gas std four
## 42 1 107 honda gas std two
## 43 0 NA isuzu gas std four
## 46 2 NA isuzu gas std two
## 47 0 145 jaguar gas std four
## 48 0 NA jaguar gas std four
## 49 0 NA jaguar gas std two
## 50 1 104 mazda gas std two
## 51 1 104 mazda gas std two
## 52 1 104 mazda gas std two
## 53 1 113 mazda gas std four
## 54 1 113 mazda gas std four
## 59 1 129 mazda gas std two
## 60 0 115 mazda gas std four
## 61 1 129 mazda gas std two
## 62 0 115 mazda gas std four
## 64 0 115 mazda gas std four
## 65 0 118 mazda gas std four
## 66 0 NA mazda diesel std four
## 67 -1 93 mercedes-benz diesel turbo four
## 68 -1 93 mercedes-benz diesel turbo four
## 69 0 93 mercedes-benz diesel turbo two
## 70 -1 93 mercedes-benz diesel turbo four
## 71 -1 NA mercedes-benz gas std four
## 72 3 142 mercedes-benz gas std two
## 73 0 NA mercedes-benz gas std four
## 74 1 NA mercedes-benz gas std two
## 75 1 NA mercury gas turbo two
## 76 2 161 mitsubishi gas std two
## 77 2 161 mitsubishi gas std two
## 78 2 161 mitsubishi gas std two
## 79 1 161 mitsubishi gas turbo two
## 80 3 153 mitsubishi gas turbo two
## 81 3 153 mitsubishi gas std two
## 82 3 NA mitsubishi gas turbo two
## 83 3 NA mitsubishi gas turbo two
## 84 3 NA mitsubishi gas turbo two
## 85 1 125 mitsubishi gas std four
## 86 1 125 mitsubishi gas std four
## 87 1 125 mitsubishi gas turbo four
## 88 -1 137 mitsubishi gas std four
## 89 1 128 nissan gas std two
## 90 1 128 nissan diesel std two
## 91 1 128 nissan gas std two
## 92 1 122 nissan gas std four
## 93 1 103 nissan gas std four
## 94 1 128 nissan gas std two
## 95 1 128 nissan gas std two
## 96 1 122 nissan gas std four
## 97 1 103 nissan gas std four
## 98 2 168 nissan gas std two
## 99 0 106 nissan gas std four
## 100 0 106 nissan gas std four
## 101 0 128 nissan gas std four
## 102 0 108 nissan gas std four
## 103 0 108 nissan gas std four
## 104 3 194 nissan gas std two
## 105 3 194 nissan gas turbo two
## 106 1 231 nissan gas std two
## 107 0 161 peugot gas std four
## 108 0 161 peugot diesel turbo four
## 109 0 NA peugot gas std four
## 110 0 NA peugot diesel turbo four
## 111 0 161 peugot gas std four
## 112 0 161 peugot diesel turbo four
## 113 0 NA peugot gas std four
## 114 0 NA peugot diesel turbo four
## 115 0 161 peugot gas std four
## 116 0 161 peugot diesel turbo four
## 117 0 161 peugot gas turbo four
## 118 1 119 plymouth gas std two
## 119 1 119 plymouth gas turbo two
## 120 1 154 plymouth gas std four
## 121 1 154 plymouth gas std four
## 122 1 154 plymouth gas std four
## 123 -1 74 plymouth gas std four
## 124 3 NA plymouth gas turbo two
## 125 3 186 porsche gas std two
## 126 3 NA porsche gas std two
## 127 3 NA porsche gas std two
## 128 3 NA porsche gas std two
## 132 3 150 saab gas std two
## 133 2 104 saab gas std four
## 134 3 150 saab gas std two
## 135 2 104 saab gas std four
## 136 3 150 saab gas turbo two
## 137 2 104 saab gas turbo four
## 138 2 83 subaru gas std two
## 139 2 83 subaru gas std two
## 140 2 83 subaru gas std two
## 141 0 102 subaru gas std four
## 142 0 102 subaru gas std four
## 143 0 102 subaru gas std four
## 144 0 102 subaru gas std four
## 145 0 102 subaru gas turbo four
## 146 0 89 subaru gas std four
## 147 0 89 subaru gas std four
## 148 0 85 subaru gas std four
## 149 0 85 subaru gas turbo four
## 150 1 87 toyota gas std two
## 151 1 87 toyota gas std two
## 152 1 74 toyota gas std four
## 153 0 77 toyota gas std four
## 154 0 81 toyota gas std four
## 155 0 91 toyota gas std four
## 156 0 91 toyota gas std four
## 157 0 91 toyota gas std four
## 158 0 91 toyota diesel std four
## 159 0 91 toyota diesel std four
## 160 0 91 toyota gas std four
## 161 0 91 toyota gas std four
## 162 0 91 toyota gas std four
## 163 1 168 toyota gas std two
## 164 1 168 toyota gas std two
## 165 1 168 toyota gas std two
## 166 1 168 toyota gas std two
## 167 2 134 toyota gas std two
## 168 2 134 toyota gas std two
## 169 2 134 toyota gas std two
## 170 2 134 toyota gas std two
## 171 2 134 toyota gas std two
## 172 2 134 toyota gas std two
## 173 -1 65 toyota gas std four
## 174 -1 65 toyota diesel turbo four
## 175 -1 65 toyota gas std four
## 176 -1 65 toyota gas std four
## 177 -1 65 toyota gas std four
## 178 3 197 toyota gas std two
## 179 3 197 toyota gas std two
## 180 -1 90 toyota gas std four
## 181 -1 NA toyota gas std four
## 182 2 122 volkswagen diesel std two
## 183 2 122 volkswagen gas std two
## 184 2 94 volkswagen diesel std four
## 185 2 94 volkswagen gas std four
## 186 2 94 volkswagen gas std four
## 187 2 94 volkswagen diesel turbo four
## 188 2 94 volkswagen gas std four
## 189 3 NA volkswagen gas std two
## 190 3 256 volkswagen gas std two
## 191 0 NA volkswagen gas std four
## 192 0 NA volkswagen diesel turbo four
## 193 0 NA volkswagen gas std four
## 194 -2 103 volvo gas std four
## 195 -1 74 volvo gas std four
## 196 -2 103 volvo gas std four
## 197 -1 74 volvo gas std four
## 198 -2 103 volvo gas turbo four
## 199 -1 74 volvo gas turbo four
## 200 -1 95 volvo gas std four
## 201 -1 95 volvo gas turbo four
## 202 -1 95 volvo gas std four
## 203 -1 95 volvo diesel turbo four
## 204 -1 95 volvo gas turbo four
## body.style drive.wheels engine.location wheel.base length width height
## 1 convertible rwd front 88.6 168.8 64.1 48.8
## 2 hatchback rwd front 94.5 171.2 65.5 52.4
## 3 sedan fwd front 99.8 176.6 66.2 54.3
## 4 sedan 4wd front 99.4 176.6 66.4 54.3
## 5 sedan fwd front 99.8 177.3 66.3 53.1
## 6 sedan fwd front 105.8 192.7 71.4 55.7
## 7 wagon fwd front 105.8 192.7 71.4 55.7
## 8 sedan fwd front 105.8 192.7 71.4 55.9
## 10 sedan rwd front 101.2 176.8 64.8 54.3
## 11 sedan rwd front 101.2 176.8 64.8 54.3
## 12 sedan rwd front 101.2 176.8 64.8 54.3
## 13 sedan rwd front 101.2 176.8 64.8 54.3
## 14 sedan rwd front 103.5 189.0 66.9 55.7
## 15 sedan rwd front 103.5 189.0 66.9 55.7
## 16 sedan rwd front 103.5 193.8 67.9 53.7
## 17 sedan rwd front 110.0 197.0 70.9 56.3
## 18 hatchback fwd front 88.4 141.1 60.3 53.2
## 19 hatchback fwd front 94.5 155.9 63.6 52.0
## 20 sedan fwd front 94.5 158.8 63.6 52.0
## 21 hatchback fwd front 93.7 157.3 63.8 50.8
## 22 hatchback fwd front 93.7 157.3 63.8 50.8
## 23 hatchback fwd front 93.7 157.3 63.8 50.8
## 24 hatchback fwd front 93.7 157.3 63.8 50.6
## 25 sedan fwd front 93.7 157.3 63.8 50.6
## 26 sedan fwd front 93.7 157.3 63.8 50.6
## 28 wagon fwd front 103.3 174.6 64.6 59.8
## 29 hatchback fwd front 95.9 173.2 66.3 50.2
## 30 hatchback fwd front 86.6 144.6 63.9 50.8
## 31 hatchback fwd front 86.6 144.6 63.9 50.8
## 32 hatchback fwd front 93.7 150.0 64.0 52.6
## 33 hatchback fwd front 93.7 150.0 64.0 52.6
## 34 hatchback fwd front 93.7 150.0 64.0 52.6
## 35 sedan fwd front 96.5 163.4 64.0 54.5
## 36 wagon fwd front 96.5 157.1 63.9 58.3
## 37 hatchback fwd front 96.5 167.5 65.2 53.3
## 38 hatchback fwd front 96.5 167.5 65.2 53.3
## 39 sedan fwd front 96.5 175.4 65.2 54.1
## 40 sedan fwd front 96.5 175.4 62.5 54.1
## 41 sedan fwd front 96.5 175.4 65.2 54.1
## 42 sedan fwd front 96.5 169.1 66.0 51.0
## 43 sedan rwd front 94.3 170.7 61.8 53.5
## 46 hatchback rwd front 96.0 172.6 65.2 51.4
## 47 sedan rwd front 113.0 199.6 69.6 52.8
## 48 sedan rwd front 113.0 199.6 69.6 52.8
## 49 sedan rwd front 102.0 191.7 70.6 47.8
## 50 hatchback fwd front 93.1 159.1 64.2 54.1
## 51 hatchback fwd front 93.1 159.1 64.2 54.1
## 52 hatchback fwd front 93.1 159.1 64.2 54.1
## 53 sedan fwd front 93.1 166.8 64.2 54.1
## 54 sedan fwd front 93.1 166.8 64.2 54.1
## 59 hatchback fwd front 98.8 177.8 66.5 53.7
## 60 sedan fwd front 98.8 177.8 66.5 55.5
## 61 hatchback fwd front 98.8 177.8 66.5 53.7
## 62 sedan fwd front 98.8 177.8 66.5 55.5
## 64 hatchback fwd front 98.8 177.8 66.5 55.5
## 65 sedan rwd front 104.9 175.0 66.1 54.4
## 66 sedan rwd front 104.9 175.0 66.1 54.4
## 67 sedan rwd front 110.0 190.9 70.3 56.5
## 68 wagon rwd front 110.0 190.9 70.3 58.7
## 69 hardtop rwd front 106.7 187.5 70.3 54.9
## 70 sedan rwd front 115.6 202.6 71.7 56.3
## 71 sedan rwd front 115.6 202.6 71.7 56.5
## 72 convertible rwd front 96.6 180.3 70.5 50.8
## 73 sedan rwd front 120.9 208.1 71.7 56.7
## 74 hardtop rwd front 112.0 199.2 72.0 55.4
## 75 hatchback rwd front 102.7 178.4 68.0 54.8
## 76 hatchback fwd front 93.7 157.3 64.4 50.8
## 77 hatchback fwd front 93.7 157.3 64.4 50.8
## 78 hatchback fwd front 93.7 157.3 64.4 50.8
## 79 hatchback fwd front 93.0 157.3 63.8 50.8
## 80 hatchback fwd front 96.3 173.0 65.4 49.4
## 81 hatchback fwd front 96.3 173.0 65.4 49.4
## 82 hatchback fwd front 95.9 173.2 66.3 50.2
## 83 hatchback fwd front 95.9 173.2 66.3 50.2
## 84 hatchback fwd front 95.9 173.2 66.3 50.2
## 85 sedan fwd front 96.3 172.4 65.4 51.6
## 86 sedan fwd front 96.3 172.4 65.4 51.6
## 87 sedan fwd front 96.3 172.4 65.4 51.6
## 88 sedan fwd front 96.3 172.4 65.4 51.6
## 89 sedan fwd front 94.5 165.3 63.8 54.5
## 90 sedan fwd front 94.5 165.3 63.8 54.5
## 91 sedan fwd front 94.5 165.3 63.8 54.5
## 92 sedan fwd front 94.5 165.3 63.8 54.5
## 93 wagon fwd front 94.5 170.2 63.8 53.5
## 94 sedan fwd front 94.5 165.3 63.8 54.5
## 95 hatchback fwd front 94.5 165.6 63.8 53.3
## 96 sedan fwd front 94.5 165.3 63.8 54.5
## 97 wagon fwd front 94.5 170.2 63.8 53.5
## 98 hardtop fwd front 95.1 162.4 63.8 53.3
## 99 hatchback fwd front 97.2 173.4 65.2 54.7
## 100 sedan fwd front 97.2 173.4 65.2 54.7
## 101 sedan fwd front 100.4 181.7 66.5 55.1
## 102 wagon fwd front 100.4 184.6 66.5 56.1
## 103 sedan fwd front 100.4 184.6 66.5 55.1
## 104 hatchback rwd front 91.3 170.7 67.9 49.7
## 105 hatchback rwd front 91.3 170.7 67.9 49.7
## 106 hatchback rwd front 99.2 178.5 67.9 49.7
## 107 sedan rwd front 107.9 186.7 68.4 56.7
## 108 sedan rwd front 107.9 186.7 68.4 56.7
## 109 wagon rwd front 114.2 198.9 68.4 58.7
## 110 wagon rwd front 114.2 198.9 68.4 58.7
## 111 sedan rwd front 107.9 186.7 68.4 56.7
## 112 sedan rwd front 107.9 186.7 68.4 56.7
## 113 wagon rwd front 114.2 198.9 68.4 56.7
## 114 wagon rwd front 114.2 198.9 68.4 58.7
## 115 sedan rwd front 107.9 186.7 68.4 56.7
## 116 sedan rwd front 107.9 186.7 68.4 56.7
## 117 sedan rwd front 108.0 186.7 68.3 56.0
## 118 hatchback fwd front 93.7 157.3 63.8 50.8
## 119 hatchback fwd front 93.7 157.3 63.8 50.8
## 120 hatchback fwd front 93.7 157.3 63.8 50.6
## 121 sedan fwd front 93.7 167.3 63.8 50.8
## 122 sedan fwd front 93.7 167.3 63.8 50.8
## 123 wagon fwd front 103.3 174.6 64.6 59.8
## 124 hatchback rwd front 95.9 173.2 66.3 50.2
## 125 hatchback rwd front 94.5 168.9 68.3 50.2
## 126 hardtop rwd rear 89.5 168.9 65.0 51.6
## 127 hardtop rwd rear 89.5 168.9 65.0 51.6
## 128 convertible rwd rear 89.5 168.9 65.0 51.6
## 132 hatchback fwd front 99.1 186.6 66.5 56.1
## 133 sedan fwd front 99.1 186.6 66.5 56.1
## 134 hatchback fwd front 99.1 186.6 66.5 56.1
## 135 sedan fwd front 99.1 186.6 66.5 56.1
## 136 hatchback fwd front 99.1 186.6 66.5 56.1
## 137 sedan fwd front 99.1 186.6 66.5 56.1
## 138 hatchback fwd front 93.7 156.9 63.4 53.7
## 139 hatchback fwd front 93.7 157.9 63.6 53.7
## 140 hatchback 4wd front 93.3 157.3 63.8 55.7
## 141 sedan fwd front 97.2 172.0 65.4 52.5
## 142 sedan fwd front 97.2 172.0 65.4 52.5
## 143 sedan fwd front 97.2 172.0 65.4 52.5
## 144 sedan 4wd front 97.0 172.0 65.4 54.3
## 145 sedan 4wd front 97.0 172.0 65.4 54.3
## 146 wagon fwd front 97.0 173.5 65.4 53.0
## 147 wagon fwd front 97.0 173.5 65.4 53.0
## 148 wagon 4wd front 96.9 173.6 65.4 54.9
## 149 wagon 4wd front 96.9 173.6 65.4 54.9
## 150 hatchback fwd front 95.7 158.7 63.6 54.5
## 151 hatchback fwd front 95.7 158.7 63.6 54.5
## 152 hatchback fwd front 95.7 158.7 63.6 54.5
## 153 wagon fwd front 95.7 169.7 63.6 59.1
## 154 wagon 4wd front 95.7 169.7 63.6 59.1
## 155 wagon 4wd front 95.7 169.7 63.6 59.1
## 156 sedan fwd front 95.7 166.3 64.4 53.0
## 157 hatchback fwd front 95.7 166.3 64.4 52.8
## 158 sedan fwd front 95.7 166.3 64.4 53.0
## 159 hatchback fwd front 95.7 166.3 64.4 52.8
## 160 sedan fwd front 95.7 166.3 64.4 53.0
## 161 hatchback fwd front 95.7 166.3 64.4 52.8
## 162 sedan fwd front 95.7 166.3 64.4 52.8
## 163 sedan rwd front 94.5 168.7 64.0 52.6
## 164 hatchback rwd front 94.5 168.7 64.0 52.6
## 165 sedan rwd front 94.5 168.7 64.0 52.6
## 166 hatchback rwd front 94.5 168.7 64.0 52.6
## 167 hardtop rwd front 98.4 176.2 65.6 52.0
## 168 hardtop rwd front 98.4 176.2 65.6 52.0
## 169 hatchback rwd front 98.4 176.2 65.6 52.0
## 170 hardtop rwd front 98.4 176.2 65.6 52.0
## 171 hatchback rwd front 98.4 176.2 65.6 52.0
## 172 convertible rwd front 98.4 176.2 65.6 53.0
## 173 sedan fwd front 102.4 175.6 66.5 54.9
## 174 sedan fwd front 102.4 175.6 66.5 54.9
## 175 hatchback fwd front 102.4 175.6 66.5 53.9
## 176 sedan fwd front 102.4 175.6 66.5 54.9
## 177 hatchback fwd front 102.4 175.6 66.5 53.9
## 178 hatchback rwd front 102.9 183.5 67.7 52.0
## 179 hatchback rwd front 102.9 183.5 67.7 52.0
## 180 sedan rwd front 104.5 187.8 66.5 54.1
## 181 wagon rwd front 104.5 187.8 66.5 54.1
## 182 sedan fwd front 97.3 171.7 65.5 55.7
## 183 sedan fwd front 97.3 171.7 65.5 55.7
## 184 sedan fwd front 97.3 171.7 65.5 55.7
## 185 sedan fwd front 97.3 171.7 65.5 55.7
## 186 sedan fwd front 97.3 171.7 65.5 55.7
## 187 sedan fwd front 97.3 171.7 65.5 55.7
## 188 sedan fwd front 97.3 171.7 65.5 55.7
## 189 convertible fwd front 94.5 159.3 64.2 55.6
## 190 hatchback fwd front 94.5 165.7 64.0 51.4
## 191 sedan fwd front 100.4 180.2 66.9 55.1
## 192 sedan fwd front 100.4 180.2 66.9 55.1
## 193 wagon fwd front 100.4 183.1 66.9 55.1
## 194 sedan rwd front 104.3 188.8 67.2 56.2
## 195 wagon rwd front 104.3 188.8 67.2 57.5
## 196 sedan rwd front 104.3 188.8 67.2 56.2
## 197 wagon rwd front 104.3 188.8 67.2 57.5
## 198 sedan rwd front 104.3 188.8 67.2 56.2
## 199 wagon rwd front 104.3 188.8 67.2 57.5
## 200 sedan rwd front 109.1 188.8 68.9 55.5
## 201 sedan rwd front 109.1 188.8 68.8 55.5
## 202 sedan rwd front 109.1 188.8 68.9 55.5
## 203 sedan rwd front 109.1 188.8 68.9 55.5
## 204 sedan rwd front 109.1 188.8 68.9 55.5
## curb.weight engine.type num.of.cylinders engine.size fuel.system bore
## 1 2548 dohc four 130 mpfi 3.47
## 2 2823 ohcv six 152 mpfi 2.68
## 3 2337 ohc four 109 mpfi 3.19
## 4 2824 ohc five 136 mpfi 3.19
## 5 2507 ohc five 136 mpfi 3.19
## 6 2844 ohc five 136 mpfi 3.19
## 7 2954 ohc five 136 mpfi 3.19
## 8 3086 ohc five 131 mpfi 3.13
## 10 2395 ohc four 108 mpfi 3.50
## 11 2395 ohc four 108 mpfi 3.50
## 12 2710 ohc six 164 mpfi 3.31
## 13 2765 ohc six 164 mpfi 3.31
## 14 3055 ohc six 164 mpfi 3.31
## 15 3230 ohc six 209 mpfi 3.62
## 16 3380 ohc six 209 mpfi 3.62
## 17 3505 ohc six 209 mpfi 3.62
## 18 1488 l three 61 2bbl 2.91
## 19 1874 ohc four 90 2bbl 3.03
## 20 1909 ohc four 90 2bbl 3.03
## 21 1876 ohc four 90 2bbl 2.97
## 22 1876 ohc four 90 2bbl 2.97
## 23 2128 ohc four 98 mpfi 3.03
## 24 1967 ohc four 90 2bbl 2.97
## 25 1989 ohc four 90 2bbl 2.97
## 26 1989 ohc four 90 2bbl 2.97
## 28 2535 ohc four 122 2bbl 3.34
## 29 2811 ohc four 156 mfi 3.60
## 30 1713 ohc four 92 1bbl 2.91
## 31 1819 ohc four 92 1bbl 2.91
## 32 1837 ohc four 79 1bbl 2.91
## 33 1940 ohc four 92 1bbl 2.91
## 34 1956 ohc four 92 1bbl 2.91
## 35 2010 ohc four 92 1bbl 2.91
## 36 2024 ohc four 92 1bbl 2.92
## 37 2236 ohc four 110 1bbl 3.15
## 38 2289 ohc four 110 1bbl 3.15
## 39 2304 ohc four 110 1bbl 3.15
## 40 2372 ohc four 110 1bbl 3.15
## 41 2465 ohc four 110 mpfi 3.15
## 42 2293 ohc four 110 2bbl 3.15
## 43 2337 ohc four 111 2bbl 3.31
## 46 2734 ohc four 119 spfi 3.43
## 47 4066 dohc six 258 mpfi 3.63
## 48 4066 dohc six 258 mpfi 3.63
## 49 3950 ohcv twelve 326 mpfi 3.54
## 50 1890 ohc four 91 2bbl 3.03
## 51 1900 ohc four 91 2bbl 3.03
## 52 1905 ohc four 91 2bbl 3.03
## 53 1945 ohc four 91 2bbl 3.03
## 54 1950 ohc four 91 2bbl 3.08
## 59 2385 ohc four 122 2bbl 3.39
## 60 2410 ohc four 122 2bbl 3.39
## 61 2385 ohc four 122 2bbl 3.39
## 62 2410 ohc four 122 2bbl 3.39
## 64 2425 ohc four 122 2bbl 3.39
## 65 2670 ohc four 140 mpfi 3.76
## 66 2700 ohc four 134 idi 3.43
## 67 3515 ohc five 183 idi 3.58
## 68 3750 ohc five 183 idi 3.58
## 69 3495 ohc five 183 idi 3.58
## 70 3770 ohc five 183 idi 3.58
## 71 3740 ohcv eight 234 mpfi 3.46
## 72 3685 ohcv eight 234 mpfi 3.46
## 73 3900 ohcv eight 308 mpfi 3.80
## 74 3715 ohcv eight 304 mpfi 3.80
## 75 2910 ohc four 140 mpfi 3.78
## 76 1918 ohc four 92 2bbl 2.97
## 77 1944 ohc four 92 2bbl 2.97
## 78 2004 ohc four 92 2bbl 2.97
## 79 2145 ohc four 98 spdi 3.03
## 80 2370 ohc four 110 spdi 3.17
## 81 2328 ohc four 122 2bbl 3.35
## 82 2833 ohc four 156 spdi 3.58
## 83 2921 ohc four 156 spdi 3.59
## 84 2926 ohc four 156 spdi 3.59
## 85 2365 ohc four 122 2bbl 3.35
## 86 2405 ohc four 122 2bbl 3.35
## 87 2403 ohc four 110 spdi 3.17
## 88 2403 ohc four 110 spdi 3.17
## 89 1889 ohc four 97 2bbl 3.15
## 90 2017 ohc four 103 idi 2.99
## 91 1918 ohc four 97 2bbl 3.15
## 92 1938 ohc four 97 2bbl 3.15
## 93 2024 ohc four 97 2bbl 3.15
## 94 1951 ohc four 97 2bbl 3.15
## 95 2028 ohc four 97 2bbl 3.15
## 96 1971 ohc four 97 2bbl 3.15
## 97 2037 ohc four 97 2bbl 3.15
## 98 2008 ohc four 97 2bbl 3.15
## 99 2324 ohc four 120 2bbl 3.33
## 100 2302 ohc four 120 2bbl 3.33
## 101 3095 ohcv six 181 mpfi 3.43
## 102 3296 ohcv six 181 mpfi 3.43
## 103 3060 ohcv six 181 mpfi 3.43
## 104 3071 ohcv six 181 mpfi 3.43
## 105 3139 ohcv six 181 mpfi 3.43
## 106 3139 ohcv six 181 mpfi 3.43
## 107 3020 l four 120 mpfi 3.46
## 108 3197 l four 152 idi 3.70
## 109 3230 l four 120 mpfi 3.46
## 110 3430 l four 152 idi 3.70
## 111 3075 l four 120 mpfi 3.46
## 112 3252 l four 152 idi 3.70
## 113 3285 l four 120 mpfi 3.46
## 114 3485 l four 152 idi 3.70
## 115 3075 l four 120 mpfi 3.46
## 116 3252 l four 152 idi 3.70
## 117 3130 l four 134 mpfi 3.61
## 118 1918 ohc four 90 2bbl 2.97
## 119 2128 ohc four 98 spdi 3.03
## 120 1967 ohc four 90 2bbl 2.97
## 121 1989 ohc four 90 2bbl 2.97
## 122 2191 ohc four 98 2bbl 2.97
## 123 2535 ohc four 122 2bbl 3.35
## 124 2818 ohc four 156 spdi 3.59
## 125 2778 ohc four 151 mpfi 3.94
## 126 2756 ohcf six 194 mpfi 3.74
## 127 2756 ohcf six 194 mpfi 3.74
## 128 2800 ohcf six 194 mpfi 3.74
## 132 2658 ohc four 121 mpfi 3.54
## 133 2695 ohc four 121 mpfi 3.54
## 134 2707 ohc four 121 mpfi 2.54
## 135 2758 ohc four 121 mpfi 3.54
## 136 2808 dohc four 121 mpfi 3.54
## 137 2847 dohc four 121 mpfi 3.54
## 138 2050 ohcf four 97 2bbl 3.62
## 139 2120 ohcf four 108 2bbl 3.62
## 140 2240 ohcf four 108 2bbl 3.62
## 141 2145 ohcf four 108 2bbl 3.62
## 142 2190 ohcf four 108 2bbl 3.62
## 143 2340 ohcf four 108 mpfi 3.62
## 144 2385 ohcf four 108 2bbl 3.62
## 145 2510 ohcf four 108 mpfi 3.62
## 146 2290 ohcf four 108 2bbl 3.62
## 147 2455 ohcf four 108 mpfi 3.62
## 148 2420 ohcf four 108 2bbl 3.62
## 149 2650 ohcf four 108 mpfi 3.62
## 150 1985 ohc four 92 2bbl 3.05
## 151 2040 ohc four 92 2bbl 3.05
## 152 2015 ohc four 92 2bbl 3.05
## 153 2280 ohc four 92 2bbl 3.05
## 154 2290 ohc four 92 2bbl 3.05
## 155 3110 ohc four 92 2bbl 3.05
## 156 2081 ohc four 98 2bbl 3.19
## 157 2109 ohc four 98 2bbl 3.19
## 158 2275 ohc four 110 idi 3.27
## 159 2275 ohc four 110 idi 3.27
## 160 2094 ohc four 98 2bbl 3.19
## 161 2122 ohc four 98 2bbl 3.19
## 162 2140 ohc four 98 2bbl 3.19
## 163 2169 ohc four 98 2bbl 3.19
## 164 2204 ohc four 98 2bbl 3.19
## 165 2265 dohc four 98 mpfi 3.24
## 166 2300 dohc four 98 mpfi 3.24
## 167 2540 ohc four 146 mpfi 3.62
## 168 2536 ohc four 146 mpfi 3.62
## 169 2551 ohc four 146 mpfi 3.62
## 170 2679 ohc four 146 mpfi 3.62
## 171 2714 ohc four 146 mpfi 3.62
## 172 2975 ohc four 146 mpfi 3.62
## 173 2326 ohc four 122 mpfi 3.31
## 174 2480 ohc four 110 idi 3.27
## 175 2414 ohc four 122 mpfi 3.31
## 176 2414 ohc four 122 mpfi 3.31
## 177 2458 ohc four 122 mpfi 3.31
## 178 2976 dohc six 171 mpfi 3.27
## 179 3016 dohc six 171 mpfi 3.27
## 180 3131 dohc six 171 mpfi 3.27
## 181 3151 dohc six 161 mpfi 3.27
## 182 2261 ohc four 97 idi 3.01
## 183 2209 ohc four 109 mpfi 3.19
## 184 2264 ohc four 97 idi 3.01
## 185 2212 ohc four 109 mpfi 3.19
## 186 2275 ohc four 109 mpfi 3.19
## 187 2319 ohc four 97 idi 3.01
## 188 2300 ohc four 109 mpfi 3.19
## 189 2254 ohc four 109 mpfi 3.19
## 190 2221 ohc four 109 mpfi 3.19
## 191 2661 ohc five 136 mpfi 3.19
## 192 2579 ohc four 97 idi 3.01
## 193 2563 ohc four 109 mpfi 3.19
## 194 2912 ohc four 141 mpfi 3.78
## 195 3034 ohc four 141 mpfi 3.78
## 196 2935 ohc four 141 mpfi 3.78
## 197 3042 ohc four 141 mpfi 3.78
## 198 3045 ohc four 130 mpfi 3.62
## 199 3157 ohc four 130 mpfi 3.62
## 200 2952 ohc four 141 mpfi 3.78
## 201 3049 ohc four 141 mpfi 3.78
## 202 3012 ohcv six 173 mpfi 3.58
## 203 3217 ohc six 145 idi 3.01
## 204 3062 ohc four 141 mpfi 3.78
## stroke compression.ratio horsepower peak.rpm city.mpg highway.mpg price
## 1 2.68 9.00 111 5000 21 27 16500
## 2 3.47 9.00 154 5000 19 26 16500
## 3 3.40 10.00 102 5500 24 30 13950
## 4 3.40 8.00 115 5500 18 22 17450
## 5 3.40 8.50 110 5500 19 25 15250
## 6 3.40 8.50 110 5500 19 25 17710
## 7 3.40 8.50 110 5500 19 25 18920
## 8 3.40 8.30 140 5500 17 20 23875
## 10 2.80 8.80 101 5800 23 29 16430
## 11 2.80 8.80 101 5800 23 29 16925
## 12 3.19 9.00 121 4250 21 28 20970
## 13 3.19 9.00 121 4250 21 28 21105
## 14 3.19 9.00 121 4250 20 25 24565
## 15 3.39 8.00 182 5400 16 22 30760
## 16 3.39 8.00 182 5400 16 22 41315
## 17 3.39 8.00 182 5400 15 20 36880
## 18 3.03 9.50 48 5100 47 53 5151
## 19 3.11 9.60 70 5400 38 43 6295
## 20 3.11 9.60 70 5400 38 43 6575
## 21 3.23 9.41 68 5500 37 41 5572
## 22 3.23 9.40 68 5500 31 38 6377
## 23 3.39 7.60 102 5500 24 30 7957
## 24 3.23 9.40 68 5500 31 38 6229
## 25 3.23 9.40 68 5500 31 38 6692
## 26 3.23 9.40 68 5500 31 38 7609
## 28 3.46 8.50 88 5000 24 30 8921
## 29 3.90 7.00 145 5000 19 24 12964
## 30 3.41 9.60 58 4800 49 54 6479
## 31 3.41 9.20 76 6000 31 38 6855
## 32 3.07 10.10 60 5500 38 42 5399
## 33 3.41 9.20 76 6000 30 34 6529
## 34 3.41 9.20 76 6000 30 34 7129
## 35 3.41 9.20 76 6000 30 34 7295
## 36 3.41 9.20 76 6000 30 34 7295
## 37 3.58 9.00 86 5800 27 33 7895
## 38 3.58 9.00 86 5800 27 33 9095
## 39 3.58 9.00 86 5800 27 33 8845
## 40 3.58 9.00 86 5800 27 33 10295
## 41 3.58 9.00 101 5800 24 28 12945
## 42 3.58 9.10 100 5500 25 31 10345
## 43 3.23 8.50 78 4800 24 29 6785
## 46 3.23 9.20 90 5000 24 29 11048
## 47 4.17 8.10 176 4750 15 19 32250
## 48 4.17 8.10 176 4750 15 19 35550
## 49 2.76 11.50 262 5000 13 17 36000
## 50 3.15 9.00 68 5000 30 31 5195
## 51 3.15 9.00 68 5000 31 38 6095
## 52 3.15 9.00 68 5000 31 38 6795
## 53 3.15 9.00 68 5000 31 38 6695
## 54 3.15 9.00 68 5000 31 38 7395
## 59 3.39 8.60 84 4800 26 32 8845
## 60 3.39 8.60 84 4800 26 32 8495
## 61 3.39 8.60 84 4800 26 32 10595
## 62 3.39 8.60 84 4800 26 32 10245
## 64 3.39 8.60 84 4800 26 32 11245
## 65 3.16 8.00 120 5000 19 27 18280
## 66 3.64 22.00 72 4200 31 39 18344
## 67 3.64 21.50 123 4350 22 25 25552
## 68 3.64 21.50 123 4350 22 25 28248
## 69 3.64 21.50 123 4350 22 25 28176
## 70 3.64 21.50 123 4350 22 25 31600
## 71 3.10 8.30 155 4750 16 18 34184
## 72 3.10 8.30 155 4750 16 18 35056
## 73 3.35 8.00 184 4500 14 16 40960
## 74 3.35 8.00 184 4500 14 16 45400
## 75 3.12 8.00 175 5000 19 24 16503
## 76 3.23 9.40 68 5500 37 41 5389
## 77 3.23 9.40 68 5500 31 38 6189
## 78 3.23 9.40 68 5500 31 38 6669
## 79 3.39 7.60 102 5500 24 30 7689
## 80 3.46 7.50 116 5500 23 30 9959
## 81 3.46 8.50 88 5000 25 32 8499
## 82 3.86 7.00 145 5000 19 24 12629
## 83 3.86 7.00 145 5000 19 24 14869
## 84 3.86 7.00 145 5000 19 24 14489
## 85 3.46 8.50 88 5000 25 32 6989
## 86 3.46 8.50 88 5000 25 32 8189
## 87 3.46 7.50 116 5500 23 30 9279
## 88 3.46 7.50 116 5500 23 30 9279
## 89 3.29 9.40 69 5200 31 37 5499
## 90 3.47 21.90 55 4800 45 50 7099
## 91 3.29 9.40 69 5200 31 37 6649
## 92 3.29 9.40 69 5200 31 37 6849
## 93 3.29 9.40 69 5200 31 37 7349
## 94 3.29 9.40 69 5200 31 37 7299
## 95 3.29 9.40 69 5200 31 37 7799
## 96 3.29 9.40 69 5200 31 37 7499
## 97 3.29 9.40 69 5200 31 37 7999
## 98 3.29 9.40 69 5200 31 37 8249
## 99 3.47 8.50 97 5200 27 34 8949
## 100 3.47 8.50 97 5200 27 34 9549
## 101 3.27 9.00 152 5200 17 22 13499
## 102 3.27 9.00 152 5200 17 22 14399
## 103 3.27 9.00 152 5200 19 25 13499
## 104 3.27 9.00 160 5200 19 25 17199
## 105 3.27 7.80 200 5200 17 23 19699
## 106 3.27 9.00 160 5200 19 25 18399
## 107 3.19 8.40 97 5000 19 24 11900
## 108 3.52 21.00 95 4150 28 33 13200
## 109 3.19 8.40 97 5000 19 24 12440
## 110 3.52 21.00 95 4150 25 25 13860
## 111 2.19 8.40 95 5000 19 24 15580
## 112 3.52 21.00 95 4150 28 33 16900
## 113 2.19 8.40 95 5000 19 24 16695
## 114 3.52 21.00 95 4150 25 25 17075
## 115 3.19 8.40 97 5000 19 24 16630
## 116 3.52 21.00 95 4150 28 33 17950
## 117 3.21 7.00 142 5600 18 24 18150
## 118 3.23 9.40 68 5500 37 41 5572
## 119 3.39 7.60 102 5500 24 30 7957
## 120 3.23 9.40 68 5500 31 38 6229
## 121 3.23 9.40 68 5500 31 38 6692
## 122 3.23 9.40 68 5500 31 38 7609
## 123 3.46 8.50 88 5000 24 30 8921
## 124 3.86 7.00 145 5000 19 24 12764
## 125 3.11 9.50 143 5500 19 27 22018
## 126 2.90 9.50 207 5900 17 25 32528
## 127 2.90 9.50 207 5900 17 25 34028
## 128 2.90 9.50 207 5900 17 25 37028
## 132 3.07 9.31 110 5250 21 28 11850
## 133 3.07 9.30 110 5250 21 28 12170
## 134 2.07 9.30 110 5250 21 28 15040
## 135 3.07 9.30 110 5250 21 28 15510
## 136 3.07 9.00 160 5500 19 26 18150
## 137 3.07 9.00 160 5500 19 26 18620
## 138 2.36 9.00 69 4900 31 36 5118
## 139 2.64 8.70 73 4400 26 31 7053
## 140 2.64 8.70 73 4400 26 31 7603
## 141 2.64 9.50 82 4800 32 37 7126
## 142 2.64 9.50 82 4400 28 33 7775
## 143 2.64 9.00 94 5200 26 32 9960
## 144 2.64 9.00 82 4800 24 25 9233
## 145 2.64 7.70 111 4800 24 29 11259
## 146 2.64 9.00 82 4800 28 32 7463
## 147 2.64 9.00 94 5200 25 31 10198
## 148 2.64 9.00 82 4800 23 29 8013
## 149 2.64 7.70 111 4800 23 23 11694
## 150 3.03 9.00 62 4800 35 39 5348
## 151 3.03 9.00 62 4800 31 38 6338
## 152 3.03 9.00 62 4800 31 38 6488
## 153 3.03 9.00 62 4800 31 37 6918
## 154 3.03 9.00 62 4800 27 32 7898
## 155 3.03 9.00 62 4800 27 32 8778
## 156 3.03 9.00 70 4800 30 37 6938
## 157 3.03 9.00 70 4800 30 37 7198
## 158 3.35 22.50 56 4500 34 36 7898
## 159 3.35 22.50 56 4500 38 47 7788
## 160 3.03 9.00 70 4800 38 47 7738
## 161 3.03 9.00 70 4800 28 34 8358
## 162 3.03 9.00 70 4800 28 34 9258
## 163 3.03 9.00 70 4800 29 34 8058
## 164 3.03 9.00 70 4800 29 34 8238
## 165 3.08 9.40 112 6600 26 29 9298
## 166 3.08 9.40 112 6600 26 29 9538
## 167 3.50 9.30 116 4800 24 30 8449
## 168 3.50 9.30 116 4800 24 30 9639
## 169 3.50 9.30 116 4800 24 30 9989
## 170 3.50 9.30 116 4800 24 30 11199
## 171 3.50 9.30 116 4800 24 30 11549
## 172 3.50 9.30 116 4800 24 30 17669
## 173 3.54 8.70 92 4200 29 34 8948
## 174 3.35 22.50 73 4500 30 33 10698
## 175 3.54 8.70 92 4200 27 32 9988
## 176 3.54 8.70 92 4200 27 32 10898
## 177 3.54 8.70 92 4200 27 32 11248
## 178 3.35 9.30 161 5200 20 24 16558
## 179 3.35 9.30 161 5200 19 24 15998
## 180 3.35 9.20 156 5200 20 24 15690
## 181 3.35 9.20 156 5200 19 24 15750
## 182 3.40 23.00 52 4800 37 46 7775
## 183 3.40 9.00 85 5250 27 34 7975
## 184 3.40 23.00 52 4800 37 46 7995
## 185 3.40 9.00 85 5250 27 34 8195
## 186 3.40 9.00 85 5250 27 34 8495
## 187 3.40 23.00 68 4500 37 42 9495
## 188 3.40 10.00 100 5500 26 32 9995
## 189 3.40 8.50 90 5500 24 29 11595
## 190 3.40 8.50 90 5500 24 29 9980
## 191 3.40 8.50 110 5500 19 24 13295
## 192 3.40 23.00 68 4500 33 38 13845
## 193 3.40 9.00 88 5500 25 31 12290
## 194 3.15 9.50 114 5400 23 28 12940
## 195 3.15 9.50 114 5400 23 28 13415
## 196 3.15 9.50 114 5400 24 28 15985
## 197 3.15 9.50 114 5400 24 28 16515
## 198 3.15 7.50 162 5100 17 22 18420
## 199 3.15 7.50 162 5100 17 22 18950
## 200 3.15 9.50 114 5400 23 28 16845
## 201 3.15 8.70 160 5300 19 25 19045
## 202 2.87 8.80 134 5500 18 23 21485
## 203 3.40 23.00 106 4800 26 27 22470
## 204 3.15 9.50 114 5400 19 25 226255.3 Fill NA values in ‘normalized.losses’ with the mean grouped by ‘symboling’
data2$normalized.losses <- ave(data2$normalized.losses, data2$symboling, FUN = function(x) {
ifelse(is.na(x), mean(x, na.rm = TRUE), x)
})
data2## symboling normalized.losses make fuel.type aspiration num.of.doors
## 1 3 174.3846 alfa-romero gas std two
## 2 1 128.1522 alfa-romero gas std two
## 3 2 164.0000 audi gas std four
## 4 2 164.0000 audi gas std four
## 5 2 125.6897 audi gas std two
## 6 1 158.0000 audi gas std four
## 7 1 128.1522 audi gas std four
## 8 1 158.0000 audi gas turbo four
## 10 2 192.0000 bmw gas std two
## 11 0 192.0000 bmw gas std four
## 12 0 188.0000 bmw gas std two
## 13 0 188.0000 bmw gas std four
## 14 1 128.1522 bmw gas std four
## 15 0 113.1667 bmw gas std four
## 16 0 113.1667 bmw gas std two
## 17 0 113.1667 bmw gas std four
## 18 2 121.0000 chevrolet gas std two
## 19 1 98.0000 chevrolet gas std two
## 20 0 81.0000 chevrolet gas std four
## 21 1 118.0000 dodge gas std two
## 22 1 118.0000 dodge gas std two
## 23 1 118.0000 dodge gas turbo two
## 24 1 148.0000 dodge gas std four
## 25 1 148.0000 dodge gas std four
## 26 1 148.0000 dodge gas std four
## 28 -1 110.0000 dodge gas std four
## 29 3 145.0000 dodge gas turbo two
## 30 2 137.0000 honda gas std two
## 31 2 137.0000 honda gas std two
## 32 1 101.0000 honda gas std two
## 33 1 101.0000 honda gas std two
## 34 1 101.0000 honda gas std two
## 35 0 110.0000 honda gas std four
## 36 0 78.0000 honda gas std four
## 37 0 106.0000 honda gas std two
## 38 0 106.0000 honda gas std two
## 39 0 85.0000 honda gas std four
## 40 0 85.0000 honda gas std four
## 41 0 85.0000 honda gas std four
## 42 1 107.0000 honda gas std two
## 43 0 113.1667 isuzu gas std four
## 46 2 125.6897 isuzu gas std two
## 47 0 145.0000 jaguar gas std four
## 48 0 113.1667 jaguar gas std four
## 49 0 113.1667 jaguar gas std two
## 50 1 104.0000 mazda gas std two
## 51 1 104.0000 mazda gas std two
## 52 1 104.0000 mazda gas std two
## 53 1 113.0000 mazda gas std four
## 54 1 113.0000 mazda gas std four
## 59 1 129.0000 mazda gas std two
## 60 0 115.0000 mazda gas std four
## 61 1 129.0000 mazda gas std two
## 62 0 115.0000 mazda gas std four
## 64 0 115.0000 mazda gas std four
## 65 0 118.0000 mazda gas std four
## 66 0 113.1667 mazda diesel std four
## 67 -1 93.0000 mercedes-benz diesel turbo four
## 68 -1 93.0000 mercedes-benz diesel turbo four
## 69 0 93.0000 mercedes-benz diesel turbo two
## 70 -1 93.0000 mercedes-benz diesel turbo four
## 71 -1 85.6000 mercedes-benz gas std four
## 72 3 142.0000 mercedes-benz gas std two
## 73 0 113.1667 mercedes-benz gas std four
## 74 1 128.1522 mercedes-benz gas std two
## 75 1 128.1522 mercury gas turbo two
## 76 2 161.0000 mitsubishi gas std two
## 77 2 161.0000 mitsubishi gas std two
## 78 2 161.0000 mitsubishi gas std two
## 79 1 161.0000 mitsubishi gas turbo two
## 80 3 153.0000 mitsubishi gas turbo two
## 81 3 153.0000 mitsubishi gas std two
## 82 3 174.3846 mitsubishi gas turbo two
## 83 3 174.3846 mitsubishi gas turbo two
## 84 3 174.3846 mitsubishi gas turbo two
## 85 1 125.0000 mitsubishi gas std four
## 86 1 125.0000 mitsubishi gas std four
## 87 1 125.0000 mitsubishi gas turbo four
## 88 -1 137.0000 mitsubishi gas std four
## 89 1 128.0000 nissan gas std two
## 90 1 128.0000 nissan diesel std two
## 91 1 128.0000 nissan gas std two
## 92 1 122.0000 nissan gas std four
## 93 1 103.0000 nissan gas std four
## 94 1 128.0000 nissan gas std two
## 95 1 128.0000 nissan gas std two
## 96 1 122.0000 nissan gas std four
## 97 1 103.0000 nissan gas std four
## 98 2 168.0000 nissan gas std two
## 99 0 106.0000 nissan gas std four
## 100 0 106.0000 nissan gas std four
## 101 0 128.0000 nissan gas std four
## 102 0 108.0000 nissan gas std four
## 103 0 108.0000 nissan gas std four
## 104 3 194.0000 nissan gas std two
## 105 3 194.0000 nissan gas turbo two
## 106 1 231.0000 nissan gas std two
## 107 0 161.0000 peugot gas std four
## 108 0 161.0000 peugot diesel turbo four
## 109 0 113.1667 peugot gas std four
## 110 0 113.1667 peugot diesel turbo four
## 111 0 161.0000 peugot gas std four
## 112 0 161.0000 peugot diesel turbo four
## 113 0 113.1667 peugot gas std four
## 114 0 113.1667 peugot diesel turbo four
## 115 0 161.0000 peugot gas std four
## 116 0 161.0000 peugot diesel turbo four
## 117 0 161.0000 peugot gas turbo four
## 118 1 119.0000 plymouth gas std two
## 119 1 119.0000 plymouth gas turbo two
## 120 1 154.0000 plymouth gas std four
## 121 1 154.0000 plymouth gas std four
## 122 1 154.0000 plymouth gas std four
## 123 -1 74.0000 plymouth gas std four
## 124 3 174.3846 plymouth gas turbo two
## 125 3 186.0000 porsche gas std two
## 126 3 174.3846 porsche gas std two
## 127 3 174.3846 porsche gas std two
## 128 3 174.3846 porsche gas std two
## 132 3 150.0000 saab gas std two
## 133 2 104.0000 saab gas std four
## 134 3 150.0000 saab gas std two
## 135 2 104.0000 saab gas std four
## 136 3 150.0000 saab gas turbo two
## 137 2 104.0000 saab gas turbo four
## 138 2 83.0000 subaru gas std two
## 139 2 83.0000 subaru gas std two
## 140 2 83.0000 subaru gas std two
## 141 0 102.0000 subaru gas std four
## 142 0 102.0000 subaru gas std four
## 143 0 102.0000 subaru gas std four
## 144 0 102.0000 subaru gas std four
## 145 0 102.0000 subaru gas turbo four
## 146 0 89.0000 subaru gas std four
## 147 0 89.0000 subaru gas std four
## 148 0 85.0000 subaru gas std four
## 149 0 85.0000 subaru gas turbo four
## 150 1 87.0000 toyota gas std two
## 151 1 87.0000 toyota gas std two
## 152 1 74.0000 toyota gas std four
## 153 0 77.0000 toyota gas std four
## 154 0 81.0000 toyota gas std four
## 155 0 91.0000 toyota gas std four
## 156 0 91.0000 toyota gas std four
## 157 0 91.0000 toyota gas std four
## 158 0 91.0000 toyota diesel std four
## 159 0 91.0000 toyota diesel std four
## 160 0 91.0000 toyota gas std four
## 161 0 91.0000 toyota gas std four
## 162 0 91.0000 toyota gas std four
## 163 1 168.0000 toyota gas std two
## 164 1 168.0000 toyota gas std two
## 165 1 168.0000 toyota gas std two
## 166 1 168.0000 toyota gas std two
## 167 2 134.0000 toyota gas std two
## 168 2 134.0000 toyota gas std two
## 169 2 134.0000 toyota gas std two
## 170 2 134.0000 toyota gas std two
## 171 2 134.0000 toyota gas std two
## 172 2 134.0000 toyota gas std two
## 173 -1 65.0000 toyota gas std four
## 174 -1 65.0000 toyota diesel turbo four
## 175 -1 65.0000 toyota gas std four
## 176 -1 65.0000 toyota gas std four
## 177 -1 65.0000 toyota gas std four
## 178 3 197.0000 toyota gas std two
## 179 3 197.0000 toyota gas std two
## 180 -1 90.0000 toyota gas std four
## 181 -1 85.6000 toyota gas std four
## 182 2 122.0000 volkswagen diesel std two
## 183 2 122.0000 volkswagen gas std two
## 184 2 94.0000 volkswagen diesel std four
## 185 2 94.0000 volkswagen gas std four
## 186 2 94.0000 volkswagen gas std four
## 187 2 94.0000 volkswagen diesel turbo four
## 188 2 94.0000 volkswagen gas std four
## 189 3 174.3846 volkswagen gas std two
## 190 3 256.0000 volkswagen gas std two
## 191 0 113.1667 volkswagen gas std four
## 192 0 113.1667 volkswagen diesel turbo four
## 193 0 113.1667 volkswagen gas std four
## 194 -2 103.0000 volvo gas std four
## 195 -1 74.0000 volvo gas std four
## 196 -2 103.0000 volvo gas std four
## 197 -1 74.0000 volvo gas std four
## 198 -2 103.0000 volvo gas turbo four
## 199 -1 74.0000 volvo gas turbo four
## 200 -1 95.0000 volvo gas std four
## 201 -1 95.0000 volvo gas turbo four
## 202 -1 95.0000 volvo gas std four
## 203 -1 95.0000 volvo diesel turbo four
## 204 -1 95.0000 volvo gas turbo four
## body.style drive.wheels engine.location wheel.base length width height
## 1 convertible rwd front 88.6 168.8 64.1 48.8
## 2 hatchback rwd front 94.5 171.2 65.5 52.4
## 3 sedan fwd front 99.8 176.6 66.2 54.3
## 4 sedan 4wd front 99.4 176.6 66.4 54.3
## 5 sedan fwd front 99.8 177.3 66.3 53.1
## 6 sedan fwd front 105.8 192.7 71.4 55.7
## 7 wagon fwd front 105.8 192.7 71.4 55.7
## 8 sedan fwd front 105.8 192.7 71.4 55.9
## 10 sedan rwd front 101.2 176.8 64.8 54.3
## 11 sedan rwd front 101.2 176.8 64.8 54.3
## 12 sedan rwd front 101.2 176.8 64.8 54.3
## 13 sedan rwd front 101.2 176.8 64.8 54.3
## 14 sedan rwd front 103.5 189.0 66.9 55.7
## 15 sedan rwd front 103.5 189.0 66.9 55.7
## 16 sedan rwd front 103.5 193.8 67.9 53.7
## 17 sedan rwd front 110.0 197.0 70.9 56.3
## 18 hatchback fwd front 88.4 141.1 60.3 53.2
## 19 hatchback fwd front 94.5 155.9 63.6 52.0
## 20 sedan fwd front 94.5 158.8 63.6 52.0
## 21 hatchback fwd front 93.7 157.3 63.8 50.8
## 22 hatchback fwd front 93.7 157.3 63.8 50.8
## 23 hatchback fwd front 93.7 157.3 63.8 50.8
## 24 hatchback fwd front 93.7 157.3 63.8 50.6
## 25 sedan fwd front 93.7 157.3 63.8 50.6
## 26 sedan fwd front 93.7 157.3 63.8 50.6
## 28 wagon fwd front 103.3 174.6 64.6 59.8
## 29 hatchback fwd front 95.9 173.2 66.3 50.2
## 30 hatchback fwd front 86.6 144.6 63.9 50.8
## 31 hatchback fwd front 86.6 144.6 63.9 50.8
## 32 hatchback fwd front 93.7 150.0 64.0 52.6
## 33 hatchback fwd front 93.7 150.0 64.0 52.6
## 34 hatchback fwd front 93.7 150.0 64.0 52.6
## 35 sedan fwd front 96.5 163.4 64.0 54.5
## 36 wagon fwd front 96.5 157.1 63.9 58.3
## 37 hatchback fwd front 96.5 167.5 65.2 53.3
## 38 hatchback fwd front 96.5 167.5 65.2 53.3
## 39 sedan fwd front 96.5 175.4 65.2 54.1
## 40 sedan fwd front 96.5 175.4 62.5 54.1
## 41 sedan fwd front 96.5 175.4 65.2 54.1
## 42 sedan fwd front 96.5 169.1 66.0 51.0
## 43 sedan rwd front 94.3 170.7 61.8 53.5
## 46 hatchback rwd front 96.0 172.6 65.2 51.4
## 47 sedan rwd front 113.0 199.6 69.6 52.8
## 48 sedan rwd front 113.0 199.6 69.6 52.8
## 49 sedan rwd front 102.0 191.7 70.6 47.8
## 50 hatchback fwd front 93.1 159.1 64.2 54.1
## 51 hatchback fwd front 93.1 159.1 64.2 54.1
## 52 hatchback fwd front 93.1 159.1 64.2 54.1
## 53 sedan fwd front 93.1 166.8 64.2 54.1
## 54 sedan fwd front 93.1 166.8 64.2 54.1
## 59 hatchback fwd front 98.8 177.8 66.5 53.7
## 60 sedan fwd front 98.8 177.8 66.5 55.5
## 61 hatchback fwd front 98.8 177.8 66.5 53.7
## 62 sedan fwd front 98.8 177.8 66.5 55.5
## 64 hatchback fwd front 98.8 177.8 66.5 55.5
## 65 sedan rwd front 104.9 175.0 66.1 54.4
## 66 sedan rwd front 104.9 175.0 66.1 54.4
## 67 sedan rwd front 110.0 190.9 70.3 56.5
## 68 wagon rwd front 110.0 190.9 70.3 58.7
## 69 hardtop rwd front 106.7 187.5 70.3 54.9
## 70 sedan rwd front 115.6 202.6 71.7 56.3
## 71 sedan rwd front 115.6 202.6 71.7 56.5
## 72 convertible rwd front 96.6 180.3 70.5 50.8
## 73 sedan rwd front 120.9 208.1 71.7 56.7
## 74 hardtop rwd front 112.0 199.2 72.0 55.4
## 75 hatchback rwd front 102.7 178.4 68.0 54.8
## 76 hatchback fwd front 93.7 157.3 64.4 50.8
## 77 hatchback fwd front 93.7 157.3 64.4 50.8
## 78 hatchback fwd front 93.7 157.3 64.4 50.8
## 79 hatchback fwd front 93.0 157.3 63.8 50.8
## 80 hatchback fwd front 96.3 173.0 65.4 49.4
## 81 hatchback fwd front 96.3 173.0 65.4 49.4
## 82 hatchback fwd front 95.9 173.2 66.3 50.2
## 83 hatchback fwd front 95.9 173.2 66.3 50.2
## 84 hatchback fwd front 95.9 173.2 66.3 50.2
## 85 sedan fwd front 96.3 172.4 65.4 51.6
## 86 sedan fwd front 96.3 172.4 65.4 51.6
## 87 sedan fwd front 96.3 172.4 65.4 51.6
## 88 sedan fwd front 96.3 172.4 65.4 51.6
## 89 sedan fwd front 94.5 165.3 63.8 54.5
## 90 sedan fwd front 94.5 165.3 63.8 54.5
## 91 sedan fwd front 94.5 165.3 63.8 54.5
## 92 sedan fwd front 94.5 165.3 63.8 54.5
## 93 wagon fwd front 94.5 170.2 63.8 53.5
## 94 sedan fwd front 94.5 165.3 63.8 54.5
## 95 hatchback fwd front 94.5 165.6 63.8 53.3
## 96 sedan fwd front 94.5 165.3 63.8 54.5
## 97 wagon fwd front 94.5 170.2 63.8 53.5
## 98 hardtop fwd front 95.1 162.4 63.8 53.3
## 99 hatchback fwd front 97.2 173.4 65.2 54.7
## 100 sedan fwd front 97.2 173.4 65.2 54.7
## 101 sedan fwd front 100.4 181.7 66.5 55.1
## 102 wagon fwd front 100.4 184.6 66.5 56.1
## 103 sedan fwd front 100.4 184.6 66.5 55.1
## 104 hatchback rwd front 91.3 170.7 67.9 49.7
## 105 hatchback rwd front 91.3 170.7 67.9 49.7
## 106 hatchback rwd front 99.2 178.5 67.9 49.7
## 107 sedan rwd front 107.9 186.7 68.4 56.7
## 108 sedan rwd front 107.9 186.7 68.4 56.7
## 109 wagon rwd front 114.2 198.9 68.4 58.7
## 110 wagon rwd front 114.2 198.9 68.4 58.7
## 111 sedan rwd front 107.9 186.7 68.4 56.7
## 112 sedan rwd front 107.9 186.7 68.4 56.7
## 113 wagon rwd front 114.2 198.9 68.4 56.7
## 114 wagon rwd front 114.2 198.9 68.4 58.7
## 115 sedan rwd front 107.9 186.7 68.4 56.7
## 116 sedan rwd front 107.9 186.7 68.4 56.7
## 117 sedan rwd front 108.0 186.7 68.3 56.0
## 118 hatchback fwd front 93.7 157.3 63.8 50.8
## 119 hatchback fwd front 93.7 157.3 63.8 50.8
## 120 hatchback fwd front 93.7 157.3 63.8 50.6
## 121 sedan fwd front 93.7 167.3 63.8 50.8
## 122 sedan fwd front 93.7 167.3 63.8 50.8
## 123 wagon fwd front 103.3 174.6 64.6 59.8
## 124 hatchback rwd front 95.9 173.2 66.3 50.2
## 125 hatchback rwd front 94.5 168.9 68.3 50.2
## 126 hardtop rwd rear 89.5 168.9 65.0 51.6
## 127 hardtop rwd rear 89.5 168.9 65.0 51.6
## 128 convertible rwd rear 89.5 168.9 65.0 51.6
## 132 hatchback fwd front 99.1 186.6 66.5 56.1
## 133 sedan fwd front 99.1 186.6 66.5 56.1
## 134 hatchback fwd front 99.1 186.6 66.5 56.1
## 135 sedan fwd front 99.1 186.6 66.5 56.1
## 136 hatchback fwd front 99.1 186.6 66.5 56.1
## 137 sedan fwd front 99.1 186.6 66.5 56.1
## 138 hatchback fwd front 93.7 156.9 63.4 53.7
## 139 hatchback fwd front 93.7 157.9 63.6 53.7
## 140 hatchback 4wd front 93.3 157.3 63.8 55.7
## 141 sedan fwd front 97.2 172.0 65.4 52.5
## 142 sedan fwd front 97.2 172.0 65.4 52.5
## 143 sedan fwd front 97.2 172.0 65.4 52.5
## 144 sedan 4wd front 97.0 172.0 65.4 54.3
## 145 sedan 4wd front 97.0 172.0 65.4 54.3
## 146 wagon fwd front 97.0 173.5 65.4 53.0
## 147 wagon fwd front 97.0 173.5 65.4 53.0
## 148 wagon 4wd front 96.9 173.6 65.4 54.9
## 149 wagon 4wd front 96.9 173.6 65.4 54.9
## 150 hatchback fwd front 95.7 158.7 63.6 54.5
## 151 hatchback fwd front 95.7 158.7 63.6 54.5
## 152 hatchback fwd front 95.7 158.7 63.6 54.5
## 153 wagon fwd front 95.7 169.7 63.6 59.1
## 154 wagon 4wd front 95.7 169.7 63.6 59.1
## 155 wagon 4wd front 95.7 169.7 63.6 59.1
## 156 sedan fwd front 95.7 166.3 64.4 53.0
## 157 hatchback fwd front 95.7 166.3 64.4 52.8
## 158 sedan fwd front 95.7 166.3 64.4 53.0
## 159 hatchback fwd front 95.7 166.3 64.4 52.8
## 160 sedan fwd front 95.7 166.3 64.4 53.0
## 161 hatchback fwd front 95.7 166.3 64.4 52.8
## 162 sedan fwd front 95.7 166.3 64.4 52.8
## 163 sedan rwd front 94.5 168.7 64.0 52.6
## 164 hatchback rwd front 94.5 168.7 64.0 52.6
## 165 sedan rwd front 94.5 168.7 64.0 52.6
## 166 hatchback rwd front 94.5 168.7 64.0 52.6
## 167 hardtop rwd front 98.4 176.2 65.6 52.0
## 168 hardtop rwd front 98.4 176.2 65.6 52.0
## 169 hatchback rwd front 98.4 176.2 65.6 52.0
## 170 hardtop rwd front 98.4 176.2 65.6 52.0
## 171 hatchback rwd front 98.4 176.2 65.6 52.0
## 172 convertible rwd front 98.4 176.2 65.6 53.0
## 173 sedan fwd front 102.4 175.6 66.5 54.9
## 174 sedan fwd front 102.4 175.6 66.5 54.9
## 175 hatchback fwd front 102.4 175.6 66.5 53.9
## 176 sedan fwd front 102.4 175.6 66.5 54.9
## 177 hatchback fwd front 102.4 175.6 66.5 53.9
## 178 hatchback rwd front 102.9 183.5 67.7 52.0
## 179 hatchback rwd front 102.9 183.5 67.7 52.0
## 180 sedan rwd front 104.5 187.8 66.5 54.1
## 181 wagon rwd front 104.5 187.8 66.5 54.1
## 182 sedan fwd front 97.3 171.7 65.5 55.7
## 183 sedan fwd front 97.3 171.7 65.5 55.7
## 184 sedan fwd front 97.3 171.7 65.5 55.7
## 185 sedan fwd front 97.3 171.7 65.5 55.7
## 186 sedan fwd front 97.3 171.7 65.5 55.7
## 187 sedan fwd front 97.3 171.7 65.5 55.7
## 188 sedan fwd front 97.3 171.7 65.5 55.7
## 189 convertible fwd front 94.5 159.3 64.2 55.6
## 190 hatchback fwd front 94.5 165.7 64.0 51.4
## 191 sedan fwd front 100.4 180.2 66.9 55.1
## 192 sedan fwd front 100.4 180.2 66.9 55.1
## 193 wagon fwd front 100.4 183.1 66.9 55.1
## 194 sedan rwd front 104.3 188.8 67.2 56.2
## 195 wagon rwd front 104.3 188.8 67.2 57.5
## 196 sedan rwd front 104.3 188.8 67.2 56.2
## 197 wagon rwd front 104.3 188.8 67.2 57.5
## 198 sedan rwd front 104.3 188.8 67.2 56.2
## 199 wagon rwd front 104.3 188.8 67.2 57.5
## 200 sedan rwd front 109.1 188.8 68.9 55.5
## 201 sedan rwd front 109.1 188.8 68.8 55.5
## 202 sedan rwd front 109.1 188.8 68.9 55.5
## 203 sedan rwd front 109.1 188.8 68.9 55.5
## 204 sedan rwd front 109.1 188.8 68.9 55.5
## curb.weight engine.type num.of.cylinders engine.size fuel.system bore
## 1 2548 dohc four 130 mpfi 3.47
## 2 2823 ohcv six 152 mpfi 2.68
## 3 2337 ohc four 109 mpfi 3.19
## 4 2824 ohc five 136 mpfi 3.19
## 5 2507 ohc five 136 mpfi 3.19
## 6 2844 ohc five 136 mpfi 3.19
## 7 2954 ohc five 136 mpfi 3.19
## 8 3086 ohc five 131 mpfi 3.13
## 10 2395 ohc four 108 mpfi 3.50
## 11 2395 ohc four 108 mpfi 3.50
## 12 2710 ohc six 164 mpfi 3.31
## 13 2765 ohc six 164 mpfi 3.31
## 14 3055 ohc six 164 mpfi 3.31
## 15 3230 ohc six 209 mpfi 3.62
## 16 3380 ohc six 209 mpfi 3.62
## 17 3505 ohc six 209 mpfi 3.62
## 18 1488 l three 61 2bbl 2.91
## 19 1874 ohc four 90 2bbl 3.03
## 20 1909 ohc four 90 2bbl 3.03
## 21 1876 ohc four 90 2bbl 2.97
## 22 1876 ohc four 90 2bbl 2.97
## 23 2128 ohc four 98 mpfi 3.03
## 24 1967 ohc four 90 2bbl 2.97
## 25 1989 ohc four 90 2bbl 2.97
## 26 1989 ohc four 90 2bbl 2.97
## 28 2535 ohc four 122 2bbl 3.34
## 29 2811 ohc four 156 mfi 3.60
## 30 1713 ohc four 92 1bbl 2.91
## 31 1819 ohc four 92 1bbl 2.91
## 32 1837 ohc four 79 1bbl 2.91
## 33 1940 ohc four 92 1bbl 2.91
## 34 1956 ohc four 92 1bbl 2.91
## 35 2010 ohc four 92 1bbl 2.91
## 36 2024 ohc four 92 1bbl 2.92
## 37 2236 ohc four 110 1bbl 3.15
## 38 2289 ohc four 110 1bbl 3.15
## 39 2304 ohc four 110 1bbl 3.15
## 40 2372 ohc four 110 1bbl 3.15
## 41 2465 ohc four 110 mpfi 3.15
## 42 2293 ohc four 110 2bbl 3.15
## 43 2337 ohc four 111 2bbl 3.31
## 46 2734 ohc four 119 spfi 3.43
## 47 4066 dohc six 258 mpfi 3.63
## 48 4066 dohc six 258 mpfi 3.63
## 49 3950 ohcv twelve 326 mpfi 3.54
## 50 1890 ohc four 91 2bbl 3.03
## 51 1900 ohc four 91 2bbl 3.03
## 52 1905 ohc four 91 2bbl 3.03
## 53 1945 ohc four 91 2bbl 3.03
## 54 1950 ohc four 91 2bbl 3.08
## 59 2385 ohc four 122 2bbl 3.39
## 60 2410 ohc four 122 2bbl 3.39
## 61 2385 ohc four 122 2bbl 3.39
## 62 2410 ohc four 122 2bbl 3.39
## 64 2425 ohc four 122 2bbl 3.39
## 65 2670 ohc four 140 mpfi 3.76
## 66 2700 ohc four 134 idi 3.43
## 67 3515 ohc five 183 idi 3.58
## 68 3750 ohc five 183 idi 3.58
## 69 3495 ohc five 183 idi 3.58
## 70 3770 ohc five 183 idi 3.58
## 71 3740 ohcv eight 234 mpfi 3.46
## 72 3685 ohcv eight 234 mpfi 3.46
## 73 3900 ohcv eight 308 mpfi 3.80
## 74 3715 ohcv eight 304 mpfi 3.80
## 75 2910 ohc four 140 mpfi 3.78
## 76 1918 ohc four 92 2bbl 2.97
## 77 1944 ohc four 92 2bbl 2.97
## 78 2004 ohc four 92 2bbl 2.97
## 79 2145 ohc four 98 spdi 3.03
## 80 2370 ohc four 110 spdi 3.17
## 81 2328 ohc four 122 2bbl 3.35
## 82 2833 ohc four 156 spdi 3.58
## 83 2921 ohc four 156 spdi 3.59
## 84 2926 ohc four 156 spdi 3.59
## 85 2365 ohc four 122 2bbl 3.35
## 86 2405 ohc four 122 2bbl 3.35
## 87 2403 ohc four 110 spdi 3.17
## 88 2403 ohc four 110 spdi 3.17
## 89 1889 ohc four 97 2bbl 3.15
## 90 2017 ohc four 103 idi 2.99
## 91 1918 ohc four 97 2bbl 3.15
## 92 1938 ohc four 97 2bbl 3.15
## 93 2024 ohc four 97 2bbl 3.15
## 94 1951 ohc four 97 2bbl 3.15
## 95 2028 ohc four 97 2bbl 3.15
## 96 1971 ohc four 97 2bbl 3.15
## 97 2037 ohc four 97 2bbl 3.15
## 98 2008 ohc four 97 2bbl 3.15
## 99 2324 ohc four 120 2bbl 3.33
## 100 2302 ohc four 120 2bbl 3.33
## 101 3095 ohcv six 181 mpfi 3.43
## 102 3296 ohcv six 181 mpfi 3.43
## 103 3060 ohcv six 181 mpfi 3.43
## 104 3071 ohcv six 181 mpfi 3.43
## 105 3139 ohcv six 181 mpfi 3.43
## 106 3139 ohcv six 181 mpfi 3.43
## 107 3020 l four 120 mpfi 3.46
## 108 3197 l four 152 idi 3.70
## 109 3230 l four 120 mpfi 3.46
## 110 3430 l four 152 idi 3.70
## 111 3075 l four 120 mpfi 3.46
## 112 3252 l four 152 idi 3.70
## 113 3285 l four 120 mpfi 3.46
## 114 3485 l four 152 idi 3.70
## 115 3075 l four 120 mpfi 3.46
## 116 3252 l four 152 idi 3.70
## 117 3130 l four 134 mpfi 3.61
## 118 1918 ohc four 90 2bbl 2.97
## 119 2128 ohc four 98 spdi 3.03
## 120 1967 ohc four 90 2bbl 2.97
## 121 1989 ohc four 90 2bbl 2.97
## 122 2191 ohc four 98 2bbl 2.97
## 123 2535 ohc four 122 2bbl 3.35
## 124 2818 ohc four 156 spdi 3.59
## 125 2778 ohc four 151 mpfi 3.94
## 126 2756 ohcf six 194 mpfi 3.74
## 127 2756 ohcf six 194 mpfi 3.74
## 128 2800 ohcf six 194 mpfi 3.74
## 132 2658 ohc four 121 mpfi 3.54
## 133 2695 ohc four 121 mpfi 3.54
## 134 2707 ohc four 121 mpfi 2.54
## 135 2758 ohc four 121 mpfi 3.54
## 136 2808 dohc four 121 mpfi 3.54
## 137 2847 dohc four 121 mpfi 3.54
## 138 2050 ohcf four 97 2bbl 3.62
## 139 2120 ohcf four 108 2bbl 3.62
## 140 2240 ohcf four 108 2bbl 3.62
## 141 2145 ohcf four 108 2bbl 3.62
## 142 2190 ohcf four 108 2bbl 3.62
## 143 2340 ohcf four 108 mpfi 3.62
## 144 2385 ohcf four 108 2bbl 3.62
## 145 2510 ohcf four 108 mpfi 3.62
## 146 2290 ohcf four 108 2bbl 3.62
## 147 2455 ohcf four 108 mpfi 3.62
## 148 2420 ohcf four 108 2bbl 3.62
## 149 2650 ohcf four 108 mpfi 3.62
## 150 1985 ohc four 92 2bbl 3.05
## 151 2040 ohc four 92 2bbl 3.05
## 152 2015 ohc four 92 2bbl 3.05
## 153 2280 ohc four 92 2bbl 3.05
## 154 2290 ohc four 92 2bbl 3.05
## 155 3110 ohc four 92 2bbl 3.05
## 156 2081 ohc four 98 2bbl 3.19
## 157 2109 ohc four 98 2bbl 3.19
## 158 2275 ohc four 110 idi 3.27
## 159 2275 ohc four 110 idi 3.27
## 160 2094 ohc four 98 2bbl 3.19
## 161 2122 ohc four 98 2bbl 3.19
## 162 2140 ohc four 98 2bbl 3.19
## 163 2169 ohc four 98 2bbl 3.19
## 164 2204 ohc four 98 2bbl 3.19
## 165 2265 dohc four 98 mpfi 3.24
## 166 2300 dohc four 98 mpfi 3.24
## 167 2540 ohc four 146 mpfi 3.62
## 168 2536 ohc four 146 mpfi 3.62
## 169 2551 ohc four 146 mpfi 3.62
## 170 2679 ohc four 146 mpfi 3.62
## 171 2714 ohc four 146 mpfi 3.62
## 172 2975 ohc four 146 mpfi 3.62
## 173 2326 ohc four 122 mpfi 3.31
## 174 2480 ohc four 110 idi 3.27
## 175 2414 ohc four 122 mpfi 3.31
## 176 2414 ohc four 122 mpfi 3.31
## 177 2458 ohc four 122 mpfi 3.31
## 178 2976 dohc six 171 mpfi 3.27
## 179 3016 dohc six 171 mpfi 3.27
## 180 3131 dohc six 171 mpfi 3.27
## 181 3151 dohc six 161 mpfi 3.27
## 182 2261 ohc four 97 idi 3.01
## 183 2209 ohc four 109 mpfi 3.19
## 184 2264 ohc four 97 idi 3.01
## 185 2212 ohc four 109 mpfi 3.19
## 186 2275 ohc four 109 mpfi 3.19
## 187 2319 ohc four 97 idi 3.01
## 188 2300 ohc four 109 mpfi 3.19
## 189 2254 ohc four 109 mpfi 3.19
## 190 2221 ohc four 109 mpfi 3.19
## 191 2661 ohc five 136 mpfi 3.19
## 192 2579 ohc four 97 idi 3.01
## 193 2563 ohc four 109 mpfi 3.19
## 194 2912 ohc four 141 mpfi 3.78
## 195 3034 ohc four 141 mpfi 3.78
## 196 2935 ohc four 141 mpfi 3.78
## 197 3042 ohc four 141 mpfi 3.78
## 198 3045 ohc four 130 mpfi 3.62
## 199 3157 ohc four 130 mpfi 3.62
## 200 2952 ohc four 141 mpfi 3.78
## 201 3049 ohc four 141 mpfi 3.78
## 202 3012 ohcv six 173 mpfi 3.58
## 203 3217 ohc six 145 idi 3.01
## 204 3062 ohc four 141 mpfi 3.78
## stroke compression.ratio horsepower peak.rpm city.mpg highway.mpg price
## 1 2.68 9.00 111 5000 21 27 16500
## 2 3.47 9.00 154 5000 19 26 16500
## 3 3.40 10.00 102 5500 24 30 13950
## 4 3.40 8.00 115 5500 18 22 17450
## 5 3.40 8.50 110 5500 19 25 15250
## 6 3.40 8.50 110 5500 19 25 17710
## 7 3.40 8.50 110 5500 19 25 18920
## 8 3.40 8.30 140 5500 17 20 23875
## 10 2.80 8.80 101 5800 23 29 16430
## 11 2.80 8.80 101 5800 23 29 16925
## 12 3.19 9.00 121 4250 21 28 20970
## 13 3.19 9.00 121 4250 21 28 21105
## 14 3.19 9.00 121 4250 20 25 24565
## 15 3.39 8.00 182 5400 16 22 30760
## 16 3.39 8.00 182 5400 16 22 41315
## 17 3.39 8.00 182 5400 15 20 36880
## 18 3.03 9.50 48 5100 47 53 5151
## 19 3.11 9.60 70 5400 38 43 6295
## 20 3.11 9.60 70 5400 38 43 6575
## 21 3.23 9.41 68 5500 37 41 5572
## 22 3.23 9.40 68 5500 31 38 6377
## 23 3.39 7.60 102 5500 24 30 7957
## 24 3.23 9.40 68 5500 31 38 6229
## 25 3.23 9.40 68 5500 31 38 6692
## 26 3.23 9.40 68 5500 31 38 7609
## 28 3.46 8.50 88 5000 24 30 8921
## 29 3.90 7.00 145 5000 19 24 12964
## 30 3.41 9.60 58 4800 49 54 6479
## 31 3.41 9.20 76 6000 31 38 6855
## 32 3.07 10.10 60 5500 38 42 5399
## 33 3.41 9.20 76 6000 30 34 6529
## 34 3.41 9.20 76 6000 30 34 7129
## 35 3.41 9.20 76 6000 30 34 7295
## 36 3.41 9.20 76 6000 30 34 7295
## 37 3.58 9.00 86 5800 27 33 7895
## 38 3.58 9.00 86 5800 27 33 9095
## 39 3.58 9.00 86 5800 27 33 8845
## 40 3.58 9.00 86 5800 27 33 10295
## 41 3.58 9.00 101 5800 24 28 12945
## 42 3.58 9.10 100 5500 25 31 10345
## 43 3.23 8.50 78 4800 24 29 6785
## 46 3.23 9.20 90 5000 24 29 11048
## 47 4.17 8.10 176 4750 15 19 32250
## 48 4.17 8.10 176 4750 15 19 35550
## 49 2.76 11.50 262 5000 13 17 36000
## 50 3.15 9.00 68 5000 30 31 5195
## 51 3.15 9.00 68 5000 31 38 6095
## 52 3.15 9.00 68 5000 31 38 6795
## 53 3.15 9.00 68 5000 31 38 6695
## 54 3.15 9.00 68 5000 31 38 7395
## 59 3.39 8.60 84 4800 26 32 8845
## 60 3.39 8.60 84 4800 26 32 8495
## 61 3.39 8.60 84 4800 26 32 10595
## 62 3.39 8.60 84 4800 26 32 10245
## 64 3.39 8.60 84 4800 26 32 11245
## 65 3.16 8.00 120 5000 19 27 18280
## 66 3.64 22.00 72 4200 31 39 18344
## 67 3.64 21.50 123 4350 22 25 25552
## 68 3.64 21.50 123 4350 22 25 28248
## 69 3.64 21.50 123 4350 22 25 28176
## 70 3.64 21.50 123 4350 22 25 31600
## 71 3.10 8.30 155 4750 16 18 34184
## 72 3.10 8.30 155 4750 16 18 35056
## 73 3.35 8.00 184 4500 14 16 40960
## 74 3.35 8.00 184 4500 14 16 45400
## 75 3.12 8.00 175 5000 19 24 16503
## 76 3.23 9.40 68 5500 37 41 5389
## 77 3.23 9.40 68 5500 31 38 6189
## 78 3.23 9.40 68 5500 31 38 6669
## 79 3.39 7.60 102 5500 24 30 7689
## 80 3.46 7.50 116 5500 23 30 9959
## 81 3.46 8.50 88 5000 25 32 8499
## 82 3.86 7.00 145 5000 19 24 12629
## 83 3.86 7.00 145 5000 19 24 14869
## 84 3.86 7.00 145 5000 19 24 14489
## 85 3.46 8.50 88 5000 25 32 6989
## 86 3.46 8.50 88 5000 25 32 8189
## 87 3.46 7.50 116 5500 23 30 9279
## 88 3.46 7.50 116 5500 23 30 9279
## 89 3.29 9.40 69 5200 31 37 5499
## 90 3.47 21.90 55 4800 45 50 7099
## 91 3.29 9.40 69 5200 31 37 6649
## 92 3.29 9.40 69 5200 31 37 6849
## 93 3.29 9.40 69 5200 31 37 7349
## 94 3.29 9.40 69 5200 31 37 7299
## 95 3.29 9.40 69 5200 31 37 7799
## 96 3.29 9.40 69 5200 31 37 7499
## 97 3.29 9.40 69 5200 31 37 7999
## 98 3.29 9.40 69 5200 31 37 8249
## 99 3.47 8.50 97 5200 27 34 8949
## 100 3.47 8.50 97 5200 27 34 9549
## 101 3.27 9.00 152 5200 17 22 13499
## 102 3.27 9.00 152 5200 17 22 14399
## 103 3.27 9.00 152 5200 19 25 13499
## 104 3.27 9.00 160 5200 19 25 17199
## 105 3.27 7.80 200 5200 17 23 19699
## 106 3.27 9.00 160 5200 19 25 18399
## 107 3.19 8.40 97 5000 19 24 11900
## 108 3.52 21.00 95 4150 28 33 13200
## 109 3.19 8.40 97 5000 19 24 12440
## 110 3.52 21.00 95 4150 25 25 13860
## 111 2.19 8.40 95 5000 19 24 15580
## 112 3.52 21.00 95 4150 28 33 16900
## 113 2.19 8.40 95 5000 19 24 16695
## 114 3.52 21.00 95 4150 25 25 17075
## 115 3.19 8.40 97 5000 19 24 16630
## 116 3.52 21.00 95 4150 28 33 17950
## 117 3.21 7.00 142 5600 18 24 18150
## 118 3.23 9.40 68 5500 37 41 5572
## 119 3.39 7.60 102 5500 24 30 7957
## 120 3.23 9.40 68 5500 31 38 6229
## 121 3.23 9.40 68 5500 31 38 6692
## 122 3.23 9.40 68 5500 31 38 7609
## 123 3.46 8.50 88 5000 24 30 8921
## 124 3.86 7.00 145 5000 19 24 12764
## 125 3.11 9.50 143 5500 19 27 22018
## 126 2.90 9.50 207 5900 17 25 32528
## 127 2.90 9.50 207 5900 17 25 34028
## 128 2.90 9.50 207 5900 17 25 37028
## 132 3.07 9.31 110 5250 21 28 11850
## 133 3.07 9.30 110 5250 21 28 12170
## 134 2.07 9.30 110 5250 21 28 15040
## 135 3.07 9.30 110 5250 21 28 15510
## 136 3.07 9.00 160 5500 19 26 18150
## 137 3.07 9.00 160 5500 19 26 18620
## 138 2.36 9.00 69 4900 31 36 5118
## 139 2.64 8.70 73 4400 26 31 7053
## 140 2.64 8.70 73 4400 26 31 7603
## 141 2.64 9.50 82 4800 32 37 7126
## 142 2.64 9.50 82 4400 28 33 7775
## 143 2.64 9.00 94 5200 26 32 9960
## 144 2.64 9.00 82 4800 24 25 9233
## 145 2.64 7.70 111 4800 24 29 11259
## 146 2.64 9.00 82 4800 28 32 7463
## 147 2.64 9.00 94 5200 25 31 10198
## 148 2.64 9.00 82 4800 23 29 8013
## 149 2.64 7.70 111 4800 23 23 11694
## 150 3.03 9.00 62 4800 35 39 5348
## 151 3.03 9.00 62 4800 31 38 6338
## 152 3.03 9.00 62 4800 31 38 6488
## 153 3.03 9.00 62 4800 31 37 6918
## 154 3.03 9.00 62 4800 27 32 7898
## 155 3.03 9.00 62 4800 27 32 8778
## 156 3.03 9.00 70 4800 30 37 6938
## 157 3.03 9.00 70 4800 30 37 7198
## 158 3.35 22.50 56 4500 34 36 7898
## 159 3.35 22.50 56 4500 38 47 7788
## 160 3.03 9.00 70 4800 38 47 7738
## 161 3.03 9.00 70 4800 28 34 8358
## 162 3.03 9.00 70 4800 28 34 9258
## 163 3.03 9.00 70 4800 29 34 8058
## 164 3.03 9.00 70 4800 29 34 8238
## 165 3.08 9.40 112 6600 26 29 9298
## 166 3.08 9.40 112 6600 26 29 9538
## 167 3.50 9.30 116 4800 24 30 8449
## 168 3.50 9.30 116 4800 24 30 9639
## 169 3.50 9.30 116 4800 24 30 9989
## 170 3.50 9.30 116 4800 24 30 11199
## 171 3.50 9.30 116 4800 24 30 11549
## 172 3.50 9.30 116 4800 24 30 17669
## 173 3.54 8.70 92 4200 29 34 8948
## 174 3.35 22.50 73 4500 30 33 10698
## 175 3.54 8.70 92 4200 27 32 9988
## 176 3.54 8.70 92 4200 27 32 10898
## 177 3.54 8.70 92 4200 27 32 11248
## 178 3.35 9.30 161 5200 20 24 16558
## 179 3.35 9.30 161 5200 19 24 15998
## 180 3.35 9.20 156 5200 20 24 15690
## 181 3.35 9.20 156 5200 19 24 15750
## 182 3.40 23.00 52 4800 37 46 7775
## 183 3.40 9.00 85 5250 27 34 7975
## 184 3.40 23.00 52 4800 37 46 7995
## 185 3.40 9.00 85 5250 27 34 8195
## 186 3.40 9.00 85 5250 27 34 8495
## 187 3.40 23.00 68 4500 37 42 9495
## 188 3.40 10.00 100 5500 26 32 9995
## 189 3.40 8.50 90 5500 24 29 11595
## 190 3.40 8.50 90 5500 24 29 9980
## 191 3.40 8.50 110 5500 19 24 13295
## 192 3.40 23.00 68 4500 33 38 13845
## 193 3.40 9.00 88 5500 25 31 12290
## 194 3.15 9.50 114 5400 23 28 12940
## 195 3.15 9.50 114 5400 23 28 13415
## 196 3.15 9.50 114 5400 24 28 15985
## 197 3.15 9.50 114 5400 24 28 16515
## 198 3.15 7.50 162 5100 17 22 18420
## 199 3.15 7.50 162 5100 17 22 18950
## 200 3.15 9.50 114 5400 23 28 16845
## 201 3.15 8.70 160 5300 19 25 19045
## 202 2.87 8.80 134 5500 18 23 21485
## 203 3.40 23.00 106 4800 26 27 22470
## 204 3.15 9.50 114 5400 19 25 226255.4 Converting Data Type
# Define the columns to convert to numeric
columns_to_convert <- c("bore", "stroke", "horsepower", "peak.rpm", "price")
# Loop through the columns and convert them to numeric
for (col in columns_to_convert) {
data2[[col]] <- as.numeric(data2[[col]])
}6 Exploratory Data Analysis (EDA)
# Filter numerical columns
numerical_columns <- data2 %>%
select(where(is.numeric)) %>%
names()
# Filter categorical columns
categorical_columns <- data2 %>%
select(where(is.character)) %>%
names()
# Print the filtered columns
print("Numerical columns:")## [1] "Numerical columns:"print(numerical_columns)## [1] "symboling" "normalized.losses" "wheel.base"
## [4] "length" "width" "height"
## [7] "curb.weight" "engine.size" "bore"
## [10] "stroke" "compression.ratio" "horsepower"
## [13] "peak.rpm" "city.mpg" "highway.mpg"
## [16] "price"print("Categorical columns:")## [1] "Categorical columns:"print(categorical_columns)## [1] "make" "fuel.type" "aspiration" "num.of.doors"
## [5] "body.style" "drive.wheels" "engine.location" "engine.type"
## [9] "num.of.cylinders" "fuel.system"6.1 Univariate Analysis
6.1.1 Frequency Plots of Categorical Features
The majority classes for each categorical feature provide valuable insights into the popular features in the car market and user preferences. In the dataset, gas fuel type, standard aspiration, four doors, sedans, fwd drive.wheels, front engine location, ohc engine type, 4 cylinders, and mpfi fuel system are predominant. This analysis highlights the popularity of certain features like front engine location (98%) and gas fuel type (90%) features, which are the highest among the majority classes determined.
In terms of car make, Toyota is the most common car make, with the highest frequency count of 32, suggesting strong popularity or market presence. Nissan & Honda also show notable frequency counts of 18 and 13, respectively, indicating they are relatively common makes in the dataset. The remaining car makes have lower frequencies, which could imply less popularity or a smaller market share in this particular sample.
# Create a list to hold the plots
plots <- list()
# Loop through each categorical column and create a plot, then add it to the list
for (col in categorical_columns[-1]) {
p <- ggplot(data2, aes_string(x = col)) +
geom_bar() +
geom_text(stat='count', aes(label=after_stat(count)), vjust=-0.5) +
ggtitle(col) +
theme_minimal() +
xlab(NULL) +
ylab(NULL) +
ylim(0, max(table(data2[[col]])) * 1.3) +
theme(axis.text.x = element_text(angle = 30, hjust = 1)) # Set y-axis limit slightly above maximum count
plots[[col]] <- p
}## Warning: `aes_string()` was deprecated in ggplot2 3.0.0.
## ℹ Please use tidy evaluation idioms with `aes()`.
## ℹ See also `vignette("ggplot2-in-packages")` for more information.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.# Arrange all the plots into a grid
grid.arrange(grobs = plots, ncol = 3,
top = textGrob("Univariate Analysis of Categorical Variables", gp = gpar(fontsize = 14)))
#For car make
ggplot(data2, aes_string(x = categorical_columns[1])) +
geom_bar() +
geom_text(stat='count', aes(label=after_stat(count)), vjust=-0.5) +
ggtitle(categorical_columns[1]) +
theme_minimal() +
xlab(NULL) +
ylab(NULL) +
ylim(0, max(table(data2[[categorical_columns[1]]])) * 1.2) + # Set y-axis limit slightly above maximum count
theme(axis.text.x = element_text(angle = 45, hjust = 1)) # Rotate x-axis labels by 90 degrees
6.1.2 Histograms for Numerical Categories
Distribution peaks can help identify common values or ranges, which is useful for understanding the typical characteristics of cars in this dataset.
Right-skewed distributions were observed for normalized losses, curb weight, engine size, horsepower, and price. This indicates that while most cars fall within a typical range, there are a few outliers with significantly higher values. Features such as engine size, horsepower, curb weight, and price are key indicators of a car’s performance and market segment.
Meanwhile, symmetric distributions that were observed for height, city MPG, and highway MPG suggest a more uniform spread around a central value, indicating consistency within these features.
# Generate summary statistics using summary() function
plots <- list()
for (col in numerical_columns) {
p <- ggplot(data2, aes(x = .data[[col]])) +
geom_histogram(bins = 15, fill = 'blue', color = 'black', alpha = 0.7) +
ggtitle(col) +
labs(x = col, y = "Frequency") +
xlab(NULL) +
ylab(NULL) +
theme_minimal()
plots[[col]] <- p
}
# Arrange all the plots into a grid
grid.arrange(grobs = plots, ncol = 4,
top = textGrob("Univariate Analysis of Numerical Variables", gp = gpar(fontsize = 14)))
6.2 Bivariate Analysis
6.2.1 Price vs Categorical Features
The box plots above reveal significant price differences associated with certain classes of categorical features.
Examples of categories associated with significantly lower prices include:
- Drive wheels with fwd and 4wd
- Three and four cylinders
- Engine type with ohc and ohcf
Fwd, four doors, and ohc engine type are not only associated with lower prices but also popular, indicating consumer preferences for affordability. This observation suggests a potential correlation between feature popularity and price points. However, without conducing formal statistical tests, it is challenging to conclusively determine significant differences in prices for the categories of other features based on graphical analysis alone.
In terms of car make, Mercedes-Benz, Jaguar, BMW and Porsche have higher median prices, suggesting their positions as premium options in the market. Meanwhile, Chevrolet, Dodge and Honda have lower median prices, indicating they are more affordable. The presence of outliers in certain brands indicates that some cars are priced significantly higher or lower than the typical range for those makes.
This analysis can help identify market positioning and pricing strategies of different car makes. It also highlights the variability in car prices within each make.
plots <- list()
for (col in categorical_columns[-1]) {
p <- ggplot(data2, aes_string(x = col, y = 'price')) +
geom_boxplot() +
ggtitle(paste(col)) +
theme_minimal() +
theme(axis.text.x = element_text(angle = 30, hjust = 1)) +
labs(x = NULL, y = "Price") # Update y-axis label
plots[[col]] <- p
}
# Arrange all the plots into a grid
grid.arrange(grobs = plots, ncol = 3,
top = textGrob("Bivariate Analysis of Price vs Categorical Variables", gp = gpar(fontsize = 14)))
ggplot(data2, aes_string(x = categorical_columns[1], y = 'price')) +
geom_boxplot() +
ggtitle(paste(categorical_columns[1])) +
theme_minimal() +
theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
labs(x = NULL, y = "Price") # Update y-axis label
6.2.2 Price vs Numerical Variables
Scatter plots provided offer a visual representation of the relationship between the price and various numerical features. Upon analysis of the plots, features such as wheel base, length, width, curb weight, engine size, and horsepower show a positive correlation with price. This indicates that larger, heavier, and more powerful cars are generally more expensive. Both city MPG and highway MPG show a negative correlation with price, suggesting that more fuel-efficient cars tend to be less expensive, potentially because high-performance cars are typically less fuel-efficient.
plots <- list()
# Bivariate analysis for numeric variables
for (col in numerical_columns) {
p <- ggplot(data2, aes_string(x = col, y = 'price')) +
geom_point() +
ggtitle(paste(col)) +
theme_minimal() +
theme(axis.text.x = element_text(angle = 0, hjust = 1)) +
theme(axis.text.y = element_text(angle = 0, hjust = 1)) + #Rotate y-axis labels horizontally
labs(x = NULL, y = NULL)
plots[[col]] <- p}
# Arrange all the plots into a grid
grid.arrange(grobs = plots, ncol = 4,
top = textGrob("Bivariate Analysis of Price vs Numerical Variables", gp = gpar(fontsize = 14)))
6.3 Correlation Analysis
Pearson’s Linear Correlation Test was conducted to test the linear relationship between all numerical features.
Engine Size, Curb Weight, Horsepower, Width, Length have strong positive correlations with price, indicating that they are key factors in determining the price of a car. Meanwhile, city MPG, and Highway MPG have strong negative correlations with price
Wheel Base and Bore have moderate positive correlations with price, indicating they are significant but not as strong as the primary predictors.
Normalized Losses, Stroke, Compression Ratio, Height, Symboling, Peak RPM have weak to very weak correlations with price, indicating they have minimal influence on the pricing of cars.
library(corrplot)## corrplot 0.92 loadedcor(data2[numerical_columns], method="pearson")## symboling normalized.losses wheel.base length
## symboling 1.00000000 0.58814292 -0.5291596 -0.36227088
## normalized.losses 0.58814292 1.00000000 -0.1563805 -0.03566146
## wheel.base -0.52915959 -0.15638053 1.0000000 0.88239411
## length -0.36227088 -0.03566146 0.8823941 1.00000000
## width -0.24237878 0.04096101 0.8186494 0.85744361
## height -0.50824126 -0.43721274 0.5837984 0.49220052
## curb.weight -0.23279091 0.08644904 0.7877767 0.88308964
## engine.size -0.06932669 0.15306926 0.5730134 0.68745892
## bore -0.15092219 0.02900189 0.5045288 0.60828824
## stroke 0.00604524 0.07278378 0.1609373 0.11878302
## compression.ratio -0.17396693 -0.14827838 0.2516232 0.15550752
## horsepower 0.06821986 0.27745856 0.3816420 0.59047323
## peak.rpm 0.23184768 0.24858595 -0.3553639 -0.27681283
## city.mpg 0.02418975 -0.24175405 -0.5148726 -0.70492158
## highway.mpg 0.09179984 -0.18683917 -0.5814205 -0.73352164
## price -0.08579408 0.16132156 0.5895412 0.69634695
## width height curb.weight engine.size bore
## symboling -0.24237878 -0.50824126 -0.23279091 -0.06932669 -0.150922188
## normalized.losses 0.04096101 -0.43721274 0.08644904 0.15306926 0.029001887
## wheel.base 0.81864939 0.58379843 0.78777668 0.57301338 0.504528795
## length 0.85744361 0.49220052 0.88308964 0.68745892 0.608288238
## width 1.00000000 0.30550276 0.86913180 0.74147318 0.545260313
## height 0.30550276 1.00000000 0.30920311 0.02773305 0.190493617
## curb.weight 0.86913180 0.30920311 1.00000000 0.85720031 0.645582794
## engine.size 0.74147318 0.02773305 0.85720031 1.00000000 0.582135321
## bore 0.54526031 0.19049362 0.64558279 0.58213532 1.000000000
## stroke 0.18270681 -0.07586437 0.17661700 0.21679871 -0.060767868
## compression.ratio 0.18775525 0.25379732 0.16102691 0.02533041 -0.003404321
## horsepower 0.62361635 -0.08049284 0.76225979 0.84536946 0.572893037
## peak.rpm -0.24903584 -0.26282579 -0.27858962 -0.21774587 -0.273415676
## city.mpg -0.66214432 -0.12023538 -0.77879121 -0.71708350 -0.600687839
## highway.mpg -0.70632311 -0.16812712 -0.81884222 -0.73800061 -0.608223920
## price 0.75615917 0.13815231 0.83537415 0.88877867 0.546601472
## stroke compression.ratio horsepower peak.rpm
## symboling 0.00604524 -0.173966933 0.06821986 0.231847683
## normalized.losses 0.07278378 -0.148278380 0.27745856 0.248585947
## wheel.base 0.16093732 0.251623187 0.38164204 -0.355363893
## length 0.11878302 0.155507520 0.59047323 -0.276812834
## width 0.18270681 0.187755250 0.62361635 -0.249035837
## height -0.07586437 0.253797316 -0.08049284 -0.262825794
## curb.weight 0.17661700 0.161026911 0.76225979 -0.278589619
## engine.size 0.21679871 0.025330415 0.84536946 -0.217745867
## bore -0.06076787 -0.003404321 0.57289304 -0.273415676
## stroke 1.00000000 0.198625944 0.10569755 -0.071044789
## compression.ratio 0.19862594 1.000000000 -0.20358465 -0.440209040
## horsepower 0.10569755 -0.203584653 1.00000000 0.101625949
## peak.rpm -0.07104479 -0.440209040 0.10162595 1.000000000
## city.mpg -0.03801297 0.314074011 -0.83400078 -0.061868727
## highway.mpg -0.04595019 0.249087955 -0.81224170 -0.009039878
## price 0.09708210 0.074537955 0.81251116 -0.103818883
## city.mpg highway.mpg price
## symboling 0.02418975 0.091799842 -0.08579408
## normalized.losses -0.24175405 -0.186839173 0.16132156
## wheel.base -0.51487263 -0.581420476 0.58954124
## length -0.70492158 -0.733521635 0.69634695
## width -0.66214432 -0.706323107 0.75615917
## height -0.12023538 -0.168127120 0.13815231
## curb.weight -0.77879121 -0.818842219 0.83537415
## engine.size -0.71708350 -0.738000612 0.88877867
## bore -0.60068784 -0.608223920 0.54660147
## stroke -0.03801297 -0.045950191 0.09708210
## compression.ratio 0.31407401 0.249087955 0.07453795
## horsepower -0.83400078 -0.812241702 0.81251116
## peak.rpm -0.06186873 -0.009039878 -0.10381888
## city.mpg 1.00000000 0.971957937 -0.70737714
## highway.mpg 0.97195794 1.000000000 -0.71968527
## price -0.70737714 -0.719685268 1.00000000relation<-cor(data2[numerical_columns])
corrplot(relation, method = "color",
addCoef.col = "black", # Color of coefficient text
number.cex = 0.5, # Size of coefficient text
tl.cex = 0.7) # Size of variable names
7 Feature Engineering
7.1 Data Transformation - Form new ‘is-high-end’ column
# Calculate the value count of price
table(data2$price)##
## 5118 5151 5195 5348 5389 5399 5499 5572 6095 6189 6229 6295 6338
## 1 1 1 1 1 1 1 2 1 1 2 1 1
## 6377 6479 6488 6529 6575 6649 6669 6692 6695 6785 6795 6849 6855
## 1 1 1 1 1 1 1 2 1 1 1 1 1
## 6918 6938 6989 7053 7099 7126 7129 7198 7295 7299 7349 7395 7463
## 1 1 1 1 1 1 1 1 2 1 1 1 1
## 7499 7603 7609 7689 7738 7775 7788 7799 7895 7898 7957 7975 7995
## 1 1 2 1 1 2 1 1 1 2 2 1 1
## 7999 8013 8058 8189 8195 8238 8249 8358 8449 8495 8499 8778 8845
## 1 1 1 1 1 1 1 1 1 2 1 1 2
## 8921 8948 8949 9095 9233 9258 9279 9298 9495 9538 9549 9639 9959
## 2 1 1 1 1 1 2 1 1 1 1 1 1
## 9960 9980 9988 9989 9995 10198 10245 10295 10345 10595 10698 10898 11048
## 1 1 1 1 1 1 1 1 1 1 1 1 1
## 11199 11245 11248 11259 11549 11595 11694 11850 11900 12170 12290 12440 12629
## 1 1 1 1 1 1 1 1 1 1 1 1 1
## 12764 12940 12945 12964 13200 13295 13415 13499 13845 13860 13950 14399 14489
## 1 1 1 1 1 1 1 2 1 1 1 1 1
## 14869 15040 15250 15510 15580 15690 15750 15985 15998 16430 16500 16503 16515
## 1 1 1 1 1 1 1 1 1 1 2 1 1
## 16558 16630 16695 16845 16900 16925 17075 17199 17450 17669 17710 17950 18150
## 1 1 1 1 1 1 1 1 1 1 1 1 2
## 18280 18344 18399 18420 18620 18920 18950 19045 19699 20970 21105 21485 22018
## 1 1 1 1 1 1 1 1 1 1 1 1 1
## 22470 22625 23875 24565 25552 28176 28248 30760 31600 32250 32528 34028 34184
## 1 1 1 1 1 1 1 1 1 1 1 1 1
## 35056 35550 36000 36880 37028 40960 41315 45400
## 1 1 1 1 1 1 1 1# Check whether there is any null value in 'price' column.
NA_price = sum(is.na(data2$price))
cat('Number of null value :',NA_price, '\n')## Number of null value : 0# Display the percentiles of price
quantile(data2$price, probs = c(0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1))## 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
## 6582.4 7295.8 7962.4 8931.8 10221.5 12553.4 15559.0 17625.2 22609.5 45400.0From, the percentiles of the price the maximum price is $45400. To classify the high end group of cars, 75th percentile of the total price is used as an indicator for the classification.
price_75th_percentile <- quantile(data2$price, 0.75)
# Classify cars based on the 75th percentile
data2$is_high_end <- ifelse(data2$price > price_75th_percentile, 1, 0)
data2[c('price', 'is_high_end')]## price is_high_end
## 1 16500 0
## 2 16500 0
## 3 13950 0
## 4 17450 1
## 5 15250 0
## 6 17710 1
## 7 18920 1
## 8 23875 1
## 10 16430 0
## 11 16925 1
## 12 20970 1
## 13 21105 1
## 14 24565 1
## 15 30760 1
## 16 41315 1
## 17 36880 1
## 18 5151 0
## 19 6295 0
## 20 6575 0
## 21 5572 0
## 22 6377 0
## 23 7957 0
## 24 6229 0
## 25 6692 0
## 26 7609 0
## 28 8921 0
## 29 12964 0
## 30 6479 0
## 31 6855 0
## 32 5399 0
## 33 6529 0
## 34 7129 0
## 35 7295 0
## 36 7295 0
## 37 7895 0
## 38 9095 0
## 39 8845 0
## 40 10295 0
## 41 12945 0
## 42 10345 0
## 43 6785 0
## 46 11048 0
## 47 32250 1
## 48 35550 1
## 49 36000 1
## 50 5195 0
## 51 6095 0
## 52 6795 0
## 53 6695 0
## 54 7395 0
## 59 8845 0
## 60 8495 0
## 61 10595 0
## 62 10245 0
## 64 11245 0
## 65 18280 1
## 66 18344 1
## 67 25552 1
## 68 28248 1
## 69 28176 1
## 70 31600 1
## 71 34184 1
## 72 35056 1
## 73 40960 1
## 74 45400 1
## 75 16503 0
## 76 5389 0
## 77 6189 0
## 78 6669 0
## 79 7689 0
## 80 9959 0
## 81 8499 0
## 82 12629 0
## 83 14869 0
## 84 14489 0
## 85 6989 0
## 86 8189 0
## 87 9279 0
## 88 9279 0
## 89 5499 0
## 90 7099 0
## 91 6649 0
## 92 6849 0
## 93 7349 0
## 94 7299 0
## 95 7799 0
## 96 7499 0
## 97 7999 0
## 98 8249 0
## 99 8949 0
## 100 9549 0
## 101 13499 0
## 102 14399 0
## 103 13499 0
## 104 17199 1
## 105 19699 1
## 106 18399 1
## 107 11900 0
## 108 13200 0
## 109 12440 0
## 110 13860 0
## 111 15580 0
## 112 16900 1
## 113 16695 1
## 114 17075 1
## 115 16630 1
## 116 17950 1
## 117 18150 1
## 118 5572 0
## 119 7957 0
## 120 6229 0
## 121 6692 0
## 122 7609 0
## 123 8921 0
## 124 12764 0
## 125 22018 1
## 126 32528 1
## 127 34028 1
## 128 37028 1
## 132 11850 0
## 133 12170 0
## 134 15040 0
## 135 15510 0
## 136 18150 1
## 137 18620 1
## 138 5118 0
## 139 7053 0
## 140 7603 0
## 141 7126 0
## 142 7775 0
## 143 9960 0
## 144 9233 0
## 145 11259 0
## 146 7463 0
## 147 10198 0
## 148 8013 0
## 149 11694 0
## 150 5348 0
## 151 6338 0
## 152 6488 0
## 153 6918 0
## 154 7898 0
## 155 8778 0
## 156 6938 0
## 157 7198 0
## 158 7898 0
## 159 7788 0
## 160 7738 0
## 161 8358 0
## 162 9258 0
## 163 8058 0
## 164 8238 0
## 165 9298 0
## 166 9538 0
## 167 8449 0
## 168 9639 0
## 169 9989 0
## 170 11199 0
## 171 11549 0
## 172 17669 1
## 173 8948 0
## 174 10698 0
## 175 9988 0
## 176 10898 0
## 177 11248 0
## 178 16558 1
## 179 15998 0
## 180 15690 0
## 181 15750 0
## 182 7775 0
## 183 7975 0
## 184 7995 0
## 185 8195 0
## 186 8495 0
## 187 9495 0
## 188 9995 0
## 189 11595 0
## 190 9980 0
## 191 13295 0
## 192 13845 0
## 193 12290 0
## 194 12940 0
## 195 13415 0
## 196 15985 0
## 197 16515 0
## 198 18420 1
## 199 18950 1
## 200 16845 1
## 201 19045 1
## 202 21485 1
## 203 22470 1
## 204 22625 17.2 Data Preparation
library(caret)## Loading required package: lattice# Select features (exclude 'price' and 'is_high_end')
target <- data2$is_high_end
regressors <- names(data2)[!names(data2) %in% c('price','is_high_end')]
features <- data2[, regressors]
# Train test split
set.seed(123)
trainIndex <- createDataPartition(target, p = .8, list = FALSE)
data_train <- data2[trainIndex,]
data_test <- data2[-trainIndex,]
# Numerical feature normalization
num <- c('symboling', 'normalized.losses', 'length','width','height', 'horsepower', 'wheel.base',
'bore', 'stroke', 'compression.ratio', 'peak.rpm','curb.weight','engine.size','city.mpg','highway.mpg')
preProcValues <- preProcess(data_train[, num], method = c("center", "scale"))
data_train[, num] <- predict(preProcValues, data_train[, num])
data_test[, num] <- predict(preProcValues, data_test[, num])
# Dummy variables for categorical data types
classes <- c('make', 'fuel.type', 'aspiration', 'num.of.doors',
'body.style', 'drive.wheels', 'engine.location',
'engine.type', 'num.of.cylinders', 'fuel.system')
dummies_train <- model.matrix(~.-1, data = data_train[, classes])
dummies_test <- model.matrix(~.-1, data = data_test[, classes])
# Combine both features
data_train <- cbind(data_train[, num], dummies_train)
data_test <- cbind(data_test[, num], dummies_test)
# Make sure the training and test sets have the same columns
common_cols <- intersect(names(data_train), names(data_test))
data_train <- data_train[, common_cols]
data_test <- data_test[, common_cols]
# Renmae column name
colnames(data_train)[colnames(data_train) == "makemercedes-benz"] <- "makemercedes.benz"
colnames(data_test)[colnames(data_test) == "makemercedes-benz"] <- "makemercedes.benz"
# Add target variable
data_train$is_high_end <- target[trainIndex]
data_test$is_high_end <- target[-trainIndex]
data_train$is_high_end<- factor(data_train$is_high_end)
data_test$is_high_end<- factor(data_test$is_high_end)
x_test <- data_test[,-ncol(data_test)]
y_test <- data_test$is_high_end8 Classifier Model
- Perform feature engineering by scaling numerical data and creating dummy variables from categorical data.
- Split the dataset into training and testing sets with an 80-20 ratio and fit the models.
- Use repeated K-fold cross-validation to evaluate the models due to the small dataset size. Compare and assess the performance of the models.
# Load required library
library(e1071) # SVM model
library(randomForest) # RF model## randomForest 4.7-1.1## Type rfNews() to see new features/changes/bug fixes.##
## Attaching package: 'randomForest'## The following object is masked from 'package:gridExtra':
##
## combine## The following object is masked from 'package:ggplot2':
##
## margin## The following object is masked from 'package:dplyr':
##
## combine# Create Support Vector Machine model
svm_model <- svm(is_high_end ~ ., data = data_train)
svm_pred <- predict(svm_model, newdata = x_test)
svm_accuracy <- mean(svm_pred == y_test)
svm_report <- confusionMatrix(svm_pred, y_test)# Create and train the Random Forest model
rf_model <- randomForest(is_high_end ~ ., data = data_train, ntree = 20)
rf_pred <- predict(rf_model, newdata = x_test)
rf_accuracy <- mean(rf_pred == y_test)
rf_report <- confusionMatrix(rf_pred, y_test)# Create and train the KNN model
knn_model <- train(is_high_end ~ ., data = data_train, method = "knn")
knn_pred <- predict(knn_model, newdata = x_test)
knn_accuracy <- mean(knn_pred == y_test)
knn_report <- confusionMatrix(knn_pred, y_test)# Print results
print(svm_accuracy)## [1] 0.8648649print(svm_report)## Confusion Matrix and Statistics
##
## Reference
## Prediction 0 1
## 0 26 3
## 1 2 6
##
## Accuracy : 0.8649
## 95% CI : (0.7123, 0.9546)
## No Information Rate : 0.7568
## P-Value [Acc > NIR] : 0.08439
##
## Kappa : 0.6186
##
## Mcnemar's Test P-Value : 1.00000
##
## Sensitivity : 0.9286
## Specificity : 0.6667
## Pos Pred Value : 0.8966
## Neg Pred Value : 0.7500
## Prevalence : 0.7568
## Detection Rate : 0.7027
## Detection Prevalence : 0.7838
## Balanced Accuracy : 0.7976
##
## 'Positive' Class : 0
## print(rf_accuracy)## [1] 0.9189189print(rf_report)## Confusion Matrix and Statistics
##
## Reference
## Prediction 0 1
## 0 26 1
## 1 2 8
##
## Accuracy : 0.9189
## 95% CI : (0.7809, 0.983)
## No Information Rate : 0.7568
## P-Value [Acc > NIR] : 0.01129
##
## Kappa : 0.7878
##
## Mcnemar's Test P-Value : 1.00000
##
## Sensitivity : 0.9286
## Specificity : 0.8889
## Pos Pred Value : 0.9630
## Neg Pred Value : 0.8000
## Prevalence : 0.7568
## Detection Rate : 0.7027
## Detection Prevalence : 0.7297
## Balanced Accuracy : 0.9087
##
## 'Positive' Class : 0
## print(knn_accuracy)## [1] 0.8918919print(knn_report)## Confusion Matrix and Statistics
##
## Reference
## Prediction 0 1
## 0 26 2
## 1 2 7
##
## Accuracy : 0.8919
## 95% CI : (0.7458, 0.9697)
## No Information Rate : 0.7568
## P-Value [Acc > NIR] : 0.0347
##
## Kappa : 0.7063
##
## Mcnemar's Test P-Value : 1.0000
##
## Sensitivity : 0.9286
## Specificity : 0.7778
## Pos Pred Value : 0.9286
## Neg Pred Value : 0.7778
## Prevalence : 0.7568
## Detection Rate : 0.7027
## Detection Prevalence : 0.7568
## Balanced Accuracy : 0.8532
##
## 'Positive' Class : 0
## 8.1 Cross Validation
The accuracy and kappa can be increase more by using cross validation. We are using repeated K fold
# CV for SVM
set.seed(50)
trctrl <- trainControl(method = "repeatedcv", number = 10, repeats = 3)
# Define the grid of hyperparameters to search
svm_radial_grid <- expand.grid(
sigma = c(0.01, 0.1, 1),
C = c(0.01, 0.1, 1, 10, 100) # cost parameter
)
svm_fit <- train(is_high_end ~ ., data = data_train, method = "svmRadial", trControl=trctrl,tuneGrid = svm_radial_grid)## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.
## Warning in .local(x, ...): Variable(s) `' constant. Cannot scale data.plot(svm_fit) 
# SVM fit graph shows sigma = 0.01 and cost = 10 are the best parameter for tunning the SVM model.8.2 CV for SVM model
library(kernlab)##
## Attaching package: 'kernlab'## The following object is masked from 'package:ggplot2':
##
## alpha#fit in the best param
svm_best_model <- ksvm(is_high_end ~ ., data = data_train, type = "C-svc", kernel = "rbfdot",
kpar = list(sigma = 0.01), C = 10)
cv_svm_pred <- predict(svm_best_model, newdata = x_test)
# Evaluate the model
cv_svm_accuracy <- mean(cv_svm_pred == y_test)
cv_svm_report <- confusionMatrix(cv_svm_pred, y_test)8.3 CV for Random Forest
# Cv for random forest
rf_grid <- expand.grid(
mtry = c(2,5,8,10,14,30)
)
rf_fit <- train(is_high_end ~ ., data = data_train, method = "rf", trControl=trctrl,tuneGrid = rf_grid)
plot(rf_fit) 
# Set up cross-validation
trctrl <- trainControl(method = "repeatedcv", number = 10, repeats = 3)
# Fit the Random Forest model using the best mtry value
rf_best_fit <- train(is_high_end ~ ., data = data_train, method = "rf",
trControl = trctrl, tuneGrid = expand.grid(mtry = 10))
# Make predictions
rf_best_pred <- predict(rf_best_fit, newdata = x_test)
# Evaluate the model
cv_rf_accuracy <- mean(rf_best_pred == y_test)
cv_rf_report <- confusionMatrix(rf_best_pred, y_test)8.4 CV for KNN model
# Set up cross-validation
trctrl <- trainControl(method = "repeatedcv", number = 10, repeats = 3)
# Define the grid of hyperparameters to search
knn_grid <- expand.grid(k = c(3, 5, 7, 9, 11, 13, 15))
# Train the k-NN model with grid search and cross-validation
knn_fit <- train(is_high_end ~ ., data = data_train, method = "knn",
trControl = trctrl, tuneGrid = knn_grid )
plot(knn_fit)
cv_knn_pred <- predict(knn_fit, newdata = x_test)
# Evaluate the model
cv_knn_accuracy <- mean(cv_knn_pred == y_test)
cv_knn_report <- confusionMatrix(cv_knn_pred, y_test)# Print results
# Compare accuracy
cat("SVM before CV:",svm_accuracy,'\n')## SVM before CV: 0.8648649cat("SVM after CV:",cv_svm_accuracy,'\n')## SVM after CV: 0.8918919cat('\n')cat("Random Forest before CV:",rf_accuracy,'\n')## Random Forest before CV: 0.9189189cat("Random Forest after CV:",cv_rf_accuracy,'\n')## Random Forest after CV: 0.9189189cat('\n')cat("KNN before CV:",knn_accuracy,'\n')## KNN before CV: 0.8918919cat("KNN after CV:",cv_knn_accuracy,'\n')## KNN after CV: 1# Show CV confusion matrix result
print(cv_svm_report)## Confusion Matrix and Statistics
##
## Reference
## Prediction 0 1
## 0 26 2
## 1 2 7
##
## Accuracy : 0.8919
## 95% CI : (0.7458, 0.9697)
## No Information Rate : 0.7568
## P-Value [Acc > NIR] : 0.0347
##
## Kappa : 0.7063
##
## Mcnemar's Test P-Value : 1.0000
##
## Sensitivity : 0.9286
## Specificity : 0.7778
## Pos Pred Value : 0.9286
## Neg Pred Value : 0.7778
## Prevalence : 0.7568
## Detection Rate : 0.7027
## Detection Prevalence : 0.7568
## Balanced Accuracy : 0.8532
##
## 'Positive' Class : 0
## print(cv_rf_report)## Confusion Matrix and Statistics
##
## Reference
## Prediction 0 1
## 0 26 1
## 1 2 8
##
## Accuracy : 0.9189
## 95% CI : (0.7809, 0.983)
## No Information Rate : 0.7568
## P-Value [Acc > NIR] : 0.01129
##
## Kappa : 0.7878
##
## Mcnemar's Test P-Value : 1.00000
##
## Sensitivity : 0.9286
## Specificity : 0.8889
## Pos Pred Value : 0.9630
## Neg Pred Value : 0.8000
## Prevalence : 0.7568
## Detection Rate : 0.7027
## Detection Prevalence : 0.7297
## Balanced Accuracy : 0.9087
##
## 'Positive' Class : 0
## print(cv_knn_report)## Confusion Matrix and Statistics
##
## Reference
## Prediction 0 1
## 0 28 0
## 1 0 9
##
## Accuracy : 1
## 95% CI : (0.9051, 1)
## No Information Rate : 0.7568
## P-Value [Acc > NIR] : 3.322e-05
##
## Kappa : 1
##
## Mcnemar's Test P-Value : NA
##
## Sensitivity : 1.0000
## Specificity : 1.0000
## Pos Pred Value : 1.0000
## Neg Pred Value : 1.0000
## Prevalence : 0.7568
## Detection Rate : 0.7568
## Detection Prevalence : 0.7568
## Balanced Accuracy : 1.0000
##
## 'Positive' Class : 0
## 9 Regression Model
Regression predicts a target attribute which is a number (as opposed to a category). Since the required libraries have been loaded and the missing values have been dealt with, we can proceed to do Linear Regression.
9.1 Linear Regression
- Linear Regression models the relationship between a dependent variable(‘price’) and one or more independent variables (‘bore’, ‘stroke’, ‘horsepower’, ‘peak.rpm’) by fitting a linear equation to observed data. In the code ‘lm’ function is used to perform linear regression on the ‘data2’ dataset.
- The ‘summary’ function provides a summary of the linear regression model, including coefficients, standards errors, t-statistics, and p-values.
- The ‘plot’ function visualizes the linear regression model by plotting the actual prices against the predicted prices and adding a regression linear
9.1.1 Load necessary libraries
library(ggplot2)
library(Metrics)##
## Attaching package: 'Metrics'## The following objects are masked from 'package:caret':
##
## precision, recall9.1.2 Linear regression
linear_model <- lm(price ~ bore + stroke + horsepower + peak.rpm + curb.weight + engine.size + compression.ratio, data = data2)9.1.3 Summary of the linear regression model
summary(linear_model)##
## Call:
## lm(formula = price ~ bore + stroke + horsepower + peak.rpm +
## curb.weight + engine.size + compression.ratio, data = data2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -12659.5 -1423.3 -159.7 1539.5 13322.4
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1.686e+04 6.591e+03 -2.558 0.011327 *
## bore -7.944e+02 1.228e+03 -0.647 0.518325
## stroke -2.969e+03 8.105e+02 -3.663 0.000326 ***
## horsepower 2.655e+01 1.574e+01 1.686 0.093463 .
## peak.rpm 2.251e+00 6.902e-01 3.261 0.001321 **
## curb.weight 3.849e+00 1.007e+00 3.820 0.000182 ***
## engine.size 1.234e+02 1.471e+01 8.387 1.28e-14 ***
## compression.ratio 2.514e+02 7.291e+01 3.448 0.000700 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3261 on 184 degrees of freedom
## Multiple R-squared: 0.8443, Adjusted R-squared: 0.8384
## F-statistic: 142.5 on 7 and 184 DF, p-value: < 2.2e-169.1.4 Predicted prices
predicted_prices <- predict(linear_model)9.1.5 Create a data frame with actual and predicted prices
results <- data.frame(actual_price = data2$price, predicted_price = predicted_prices)9.1.6 Determine the range of prices
price_range <- range(data2$price, predicted_prices)9.1.7 Visualize the linear regression model
ggplot(results, aes(x = actual_price, y = predicted_price)) +
geom_point(color = "blue") + # Scatterplot of actual vs predicted price
geom_abline(intercept = coef(linear_model)[1], slope = coef(linear_model)[2], color = "red") + # Add regression line
labs(title = "Linear Regression", x = "Actual Price", y = "Predicted Price") + # Labels and title
xlim(price_range) + # Limit x-axis to the range of prices
ylim(price_range) # Limit y-axis to the range of prices
9.1.8 Assess price volatility and risk
ggplot(data2, aes(x = predicted_prices, fill = is_high_end)) +
geom_density(alpha = 0.5) +
ggtitle("Distribution of Predicted Prices by Risk Level") +
xlab("Predicted Price") +
ylab("Density")## Warning: The following aesthetics were dropped during statistical transformation: fill.
## ℹ This can happen when ggplot fails to infer the correct grouping structure in
## the data.
## ℹ Did you forget to specify a `group` aesthetic or to convert a numerical
## variable into a factor?
9.1.9 Results Interpretation
It’s already a good result: R squared of 0.83 and the regression line in the scatter plot is straight and aligned with the diagonal. This indicates that the model’s predictions are very close to the actual prices. The statistical metrics show low residual standard error, high R-squared values, and a significant F-statistic, it suggests that the model is performing well in predicting the prices. Furthermore, there is a strong linear relationship between the predicted and actual prices.
9.1.10 Evaluate the model
MAE and MSE is a metric to evaluate performance of regression models by measuring the average absolute difference between the actual and predicted values. In the code, the ‘MAE’ function from the ‘Metrics’ library is used to compute the MAE for the linear regression model.
9.1.11 First Trial: Making predictions on the test set
y_test <- data2$price
y_pred <- predict(linear_model, newdata = data2)9.1.12 Mean Squared Error (MSE)
mse <- mean(linear_model$residuals^2)
cat("Mean Squared Error (MSE):", mse, "\n")## Mean Squared Error (MSE): 101885739.1.13 Calculate R-squared (R2)
r2 <- cor(y_test, y_pred)^ 2
cat("R-squared (R2):", r2, "\n")## R-squared (R2): 0.84429019.1.14 Mean Absolute Error (MAE)
mae <- MAE(data2$price, predict(linear_model))
cat("Mean Absolute Error (MAE):", mae, "\n")## Mean Absolute Error (MAE): 2240.889###Second Trial
MSE <- function(y_true, y_pred) {
mse <- mean((y_true - y_pred) ^ 2)
cat('MSE:', round(mse, 3), "\n")
return(mse)
}MSE(y_test, y_pred)## MSE: 10188573## [1] 10188573R2 <- function(y_true, y_pred) {
r2 <- cor(y_true, y_pred) ^ 2
cat('R2:', round(r2, 3), "\n")
return(r2)
}R2(y_test, y_pred)## R2: 0.844## [1] 0.84429019.2 For the Lasso Model
Lasso regression is a linear regression technique that performs both variable selection and regularization to improve the model’s prediction accuracy and interpretability. It penalizes the absolute size of the coefficients to encourage sparse solutions.
9.2.1 Load necessary libraries
library(glmnet)## Loading required package: Matrix## Loaded glmnet 4.1-8library(ggplot2)
library(Metrics)9.2.2 Fit Lasso regression model
lasso_model <- glmnet(x = as.matrix(data2[, c("bore", "stroke", "horsepower", "peak.rpm")]),
y = data2$price,
alpha = 1) # Lasso regression (alpha = 1)9.2.3 Summary of the Lasso regression model
print(lasso_model)##
## Call: glmnet(x = as.matrix(data2[, c("bore", "stroke", "horsepower", "peak.rpm")]), y = data2$price, alpha = 1)
##
## Df %Dev Lambda
## 1 0 0.00 6572.0
## 2 1 11.21 5989.0
## 3 1 20.51 5457.0
## 4 1 28.24 4972.0
## 5 1 34.65 4530.0
## 6 1 39.98 4128.0
## 7 1 44.40 3761.0
## 8 1 48.07 3427.0
## 9 1 51.12 3122.0
## 10 1 53.65 2845.0
## 11 1 55.75 2592.0
## 12 1 57.49 2362.0
## 13 1 58.94 2152.0
## 14 1 60.14 1961.0
## 15 1 61.14 1787.0
## 16 1 61.97 1628.0
## 17 2 62.72 1483.0
## 18 2 63.45 1352.0
## 19 3 64.42 1232.0
## 20 3 65.29 1122.0
## 21 3 66.02 1022.0
## 22 3 66.63 931.6
## 23 3 67.13 848.9
## 24 3 67.55 773.5
## 25 3 67.90 704.7
## 26 3 68.18 642.1
## 27 3 68.42 585.1
## 28 3 68.62 533.1
## 29 3 68.79 485.8
## 30 3 68.92 442.6
## 31 3 69.04 403.3
## 32 3 69.13 367.5
## 33 3 69.21 334.8
## 34 3 69.27 305.1
## 35 3 69.33 278.0
## 36 3 69.37 253.3
## 37 3 69.41 230.8
## 38 3 69.44 210.3
## 39 3 69.46 191.6
## 40 3 69.49 174.6
## 41 3 69.50 159.1
## 42 3 69.52 144.9
## 43 3 69.53 132.1
## 44 3 69.54 120.3
## 45 3 69.55 109.6
## 46 3 69.56 99.9
## 47 3 69.56 91.0
## 48 3 69.57 82.9
## 49 3 69.57 75.6
## 50 3 69.57 68.8
## 51 3 69.58 62.7
## 52 3 69.58 57.2
## 53 3 69.58 52.1
## 54 3 69.58 47.5
## 55 3 69.58 43.2
## 56 3 69.58 39.4
## 57 3 69.59 35.9
## 58 3 69.59 32.7
## 59 3 69.59 29.89.2.4 Visualize the coefficients
plot(lasso_model, xvar = "lambda", label = TRUE)
9.2.5 Evaluate the Lasso model
9.2.6 Predict on training data
lambda_min <- lasso_model$lambda.min # Get lambda.min value from the model object
predicted <- predict(lasso_model, newx = as.matrix(data2[, c("bore", "stroke", "horsepower", "peak.rpm")]), s = lambda_min)9.2.7 Check if ‘predicted’ is null
if (is.null(predicted)) {
cat("Error: 'predicted' is null. Check predict function call.\n")
} else {
# Mean Absolute Error (MAE)
mae <- MAE(data2$price, predicted)
cat("Mean Absolute Error (MAE):", mae, "\n")
# Mean Squared Error (MSE)
mse <- mean((data2$price - predicted)^2)
cat("Mean Squared Error (MSE):", mse, "\n")
}## Mean Absolute Error (MAE): 3572.318
## Mean Squared Error (MSE): 25106915###Third Trial ### Define the MSE function
MSE <- function(y_true, y_pred) {
mse <- mean((y_true - y_pred) ^ 2)
cat('MSE:', round(mse, 3), "\n")
return(mse)
}9.2.8 Define the R2 function
R2 <- function(y_true, y_pred) {
r2 <- cor(y_true, y_pred) ^ 2
cat('R2:', round(r2, 3), "\n")
return(r2)
}9.2.9 Calculate MSE and R2
MSE(y_test, y_pred)## MSE: 10188573## [1] 10188573R2(y_test, y_pred)## R2: 0.844## [1] 0.84429019.2.10 Results Interpretation
It’s already a good result: R squared of 0.83-0.86 and the regression line in the scatter plot is straight and aligned with the diagonal it indicates that the model’s predictions are very close to the actual prices. The statistical metrics show low residual standard error, high R-squared values, and a significant F-statistic, it suggest that the model is performing well in predicting the prices. And there is a strong linear relationship between the predicted and actual prices.
9.2.11 Make predictions on the training set
train_preds <- predict(linear_model, newdata = data2)9.2.12 Create a data frame with predictions and true values
preds <- data.frame(preds = train_preds, true = data2$price)
preds$residuals <- preds$true - preds$preds9.2.13 Plot residuals
ggplot(preds, aes(x = preds, y = residuals)) +
geom_point(color = "blue") +
labs(title = "Residual Plot",
x = "Predicted Values",
y = "Residuals") +
theme_minimal() +
theme(plot.title = element_text(hjust = 0.5)) # Center the plot title
### For the Lasso Model Lasso regression is a linear regression
technique that performs both variable selection and regularization to
improve the model’s prediction accuracy and interpretability. It
penalizes the absolute size of the coefficients to encourage sparse
solutions.
9.2.14 Load necessary libraries
library(glmnet)
library(ggplot2)
library(Metrics)9.2.15 Fit Lasso regression model
lasso_model <- glmnet(x = as.matrix(data2[, c("bore", "stroke", "horsepower", "peak.rpm")]),
y = data2$price,
alpha = 1) # Lasso regression (alpha = 1)9.2.16 Summary of the Lasso regression model
print(lasso_model)##
## Call: glmnet(x = as.matrix(data2[, c("bore", "stroke", "horsepower", "peak.rpm")]), y = data2$price, alpha = 1)
##
## Df %Dev Lambda
## 1 0 0.00 6572.0
## 2 1 11.21 5989.0
## 3 1 20.51 5457.0
## 4 1 28.24 4972.0
## 5 1 34.65 4530.0
## 6 1 39.98 4128.0
## 7 1 44.40 3761.0
## 8 1 48.07 3427.0
## 9 1 51.12 3122.0
## 10 1 53.65 2845.0
## 11 1 55.75 2592.0
## 12 1 57.49 2362.0
## 13 1 58.94 2152.0
## 14 1 60.14 1961.0
## 15 1 61.14 1787.0
## 16 1 61.97 1628.0
## 17 2 62.72 1483.0
## 18 2 63.45 1352.0
## 19 3 64.42 1232.0
## 20 3 65.29 1122.0
## 21 3 66.02 1022.0
## 22 3 66.63 931.6
## 23 3 67.13 848.9
## 24 3 67.55 773.5
## 25 3 67.90 704.7
## 26 3 68.18 642.1
## 27 3 68.42 585.1
## 28 3 68.62 533.1
## 29 3 68.79 485.8
## 30 3 68.92 442.6
## 31 3 69.04 403.3
## 32 3 69.13 367.5
## 33 3 69.21 334.8
## 34 3 69.27 305.1
## 35 3 69.33 278.0
## 36 3 69.37 253.3
## 37 3 69.41 230.8
## 38 3 69.44 210.3
## 39 3 69.46 191.6
## 40 3 69.49 174.6
## 41 3 69.50 159.1
## 42 3 69.52 144.9
## 43 3 69.53 132.1
## 44 3 69.54 120.3
## 45 3 69.55 109.6
## 46 3 69.56 99.9
## 47 3 69.56 91.0
## 48 3 69.57 82.9
## 49 3 69.57 75.6
## 50 3 69.57 68.8
## 51 3 69.58 62.7
## 52 3 69.58 57.2
## 53 3 69.58 52.1
## 54 3 69.58 47.5
## 55 3 69.58 43.2
## 56 3 69.58 39.4
## 57 3 69.59 35.9
## 58 3 69.59 32.7
## 59 3 69.59 29.8library(glmnet)
X <- as.matrix(data2[, -ncol(data2)]) # All columns except the last one
y <- data2[, ncol(data2)] # The last column as the response variable
set.seed(123) # Set seed for reproducibility
lassocv <- cv.glmnet(X, y, alpha = 1)## Warning in storage.mode(xd) <- "double": NAs introduced by coercion
## Warning in storage.mode(xd) <- "double": NAs introduced by coercion
## Warning in storage.mode(xd) <- "double": NAs introduced by coercion
## Warning in storage.mode(xd) <- "double": NAs introduced by coercion
## Warning in storage.mode(xd) <- "double": NAs introduced by coercion
## Warning in storage.mode(xd) <- "double": NAs introduced by coercion
## Warning in storage.mode(xd) <- "double": NAs introduced by coercion
## Warning in storage.mode(xd) <- "double": NAs introduced by coercion
## Warning in storage.mode(xd) <- "double": NAs introduced by coercion
## Warning in storage.mode(xd) <- "double": NAs introduced by coercion
## Warning in storage.mode(xd) <- "double": NAs introduced by coercion## Warning in cbind2(1, newx) %*% nbeta: NAs introduced by coercion
## Warning in cbind2(1, newx) %*% nbeta: NAs introduced by coercion
## Warning in cbind2(1, newx) %*% nbeta: NAs introduced by coercion
## Warning in cbind2(1, newx) %*% nbeta: NAs introduced by coercion
## Warning in cbind2(1, newx) %*% nbeta: NAs introduced by coercion
## Warning in cbind2(1, newx) %*% nbeta: NAs introduced by coercion
## Warning in cbind2(1, newx) %*% nbeta: NAs introduced by coercion
## Warning in cbind2(1, newx) %*% nbeta: NAs introduced by coercion
## Warning in cbind2(1, newx) %*% nbeta: NAs introduced by coercion
## Warning in cbind2(1, newx) %*% nbeta: NAs introduced by coercion# Get the best lambda value
best_lambda <- lassocv$lambda.min
# Print the best lambda value
print(best_lambda)## [1] 0.001187405coefs <- as.matrix(coef(lassocv, s = "lambda.min"))
coefs <- coefs[-1, ] # Remove the intercept
# prints out the number of picked/eliminated features
cat("Lasso picked ", sum(coefs != 0), " features and eliminated the other ",
sum(coefs == 0), " features.\n")## Lasso picked 16 features and eliminated the other 10 features.# Create a data frame of coefficients and feature names
coefs_df <- data.frame(coef = coefs, feature = colnames(X))
# Print the number of picked/eliminated features
num_picked <- sum(coefs != 0)
num_eliminated <- sum(coefs == 0)
cat("Lasso picked ", num_picked, " features and eliminated the other ", num_eliminated, " features.\n")## Lasso picked 16 features and eliminated the other 10 features.# Take the first and last 5 coefficients for plotting
top_bottom_coefs <- rbind(head(coefs_df[order(coefs_df$coef),], 5),
tail(coefs_df[order(coefs_df$coef),], 5))
# Plot the coefficients
barplot(top_bottom_coefs$coef, horiz = TRUE, names.arg = top_bottom_coefs$feature,
main = "Coefficients in the Lasso Model", col = "steelblue")
# Assuming 'data2' is already loaded into the environment
# Example: data2 <- read.csv('path_to_your_csv_file.csv')
# Define the target variable and feature dataframe
target_column_name <- "price" # The target column name
if (!target_column_name %in% colnames(data2)) {
stop("The target column 'price' is not found in the dataframe")
}
target <- data2[[target_column_name]] # Ensure this column exists in your dataframe
features <- data2[, !names(data2) %in% target_column_name]# Replace missing values with median
features <- na.omit(features)
# Categorical variables
classes <- c('make', 'fuel.type', 'aspiration', 'num.of.doors', 'body.style', 'drive.wheels', 'engine.location', 'engine.type', 'num.of.cylinders', 'fuel.system')# Check if the classes exist in the features dataframe and print missing variables
missing_vars <- setdiff(classes, colnames(features))
if (length(missing_vars) > 0) {
stop(paste("The following categorical variables are not found in the features dataframe:", paste(missing_vars, collapse = ", ")))
}# Create dummy variables for categorical columns
dummies <- model.matrix(~ . - 1, data = features[, classes, drop = FALSE])
features <- cbind(features, dummies)
features <- features[, !colnames(features) %in% classes]
# Print dataset dimensions and first few rows
print(paste("In total:", dim(features)))## [1] "In total: 192" "In total: 62"head(features)## symboling normalized.losses wheel.base length width height curb.weight
## 1 3 174.3846 88.6 168.8 64.1 48.8 2548
## 2 1 128.1522 94.5 171.2 65.5 52.4 2823
## 3 2 164.0000 99.8 176.6 66.2 54.3 2337
## 4 2 164.0000 99.4 176.6 66.4 54.3 2824
## 5 2 125.6897 99.8 177.3 66.3 53.1 2507
## 6 1 158.0000 105.8 192.7 71.4 55.7 2844
## engine.size bore stroke compression.ratio horsepower peak.rpm city.mpg
## 1 130 3.47 2.68 9.0 111 5000 21
## 2 152 2.68 3.47 9.0 154 5000 19
## 3 109 3.19 3.40 10.0 102 5500 24
## 4 136 3.19 3.40 8.0 115 5500 18
## 5 136 3.19 3.40 8.5 110 5500 19
## 6 136 3.19 3.40 8.5 110 5500 19
## highway.mpg is_high_end makealfa-romero makeaudi makebmw makechevrolet
## 1 27 0 1 0 0 0
## 2 26 0 1 0 0 0
## 3 30 0 0 1 0 0
## 4 22 1 0 1 0 0
## 5 25 0 0 1 0 0
## 6 25 1 0 1 0 0
## makedodge makehonda makeisuzu makejaguar makemazda makemercedes-benz
## 1 0 0 0 0 0 0
## 2 0 0 0 0 0 0
## 3 0 0 0 0 0 0
## 4 0 0 0 0 0 0
## 5 0 0 0 0 0 0
## 6 0 0 0 0 0 0
## makemercury makemitsubishi makenissan makepeugot makeplymouth makeporsche
## 1 0 0 0 0 0 0
## 2 0 0 0 0 0 0
## 3 0 0 0 0 0 0
## 4 0 0 0 0 0 0
## 5 0 0 0 0 0 0
## 6 0 0 0 0 0 0
## makesaab makesubaru maketoyota makevolkswagen makevolvo fuel.typegas
## 1 0 0 0 0 0 1
## 2 0 0 0 0 0 1
## 3 0 0 0 0 0 1
## 4 0 0 0 0 0 1
## 5 0 0 0 0 0 1
## 6 0 0 0 0 0 1
## aspirationturbo num.of.doorstwo body.stylehardtop body.stylehatchback
## 1 0 1 0 0
## 2 0 1 0 1
## 3 0 0 0 0
## 4 0 0 0 0
## 5 0 1 0 0
## 6 0 0 0 0
## body.stylesedan body.stylewagon drive.wheelsfwd drive.wheelsrwd
## 1 0 0 0 1
## 2 0 0 0 1
## 3 1 0 1 0
## 4 1 0 0 0
## 5 1 0 1 0
## 6 1 0 1 0
## engine.locationrear engine.typel engine.typeohc engine.typeohcf
## 1 0 0 0 0
## 2 0 0 0 0
## 3 0 0 1 0
## 4 0 0 1 0
## 5 0 0 1 0
## 6 0 0 1 0
## engine.typeohcv num.of.cylindersfive num.of.cylindersfour num.of.cylinderssix
## 1 0 0 1 0
## 2 1 0 0 1
## 3 0 0 1 0
## 4 0 1 0 0
## 5 0 1 0 0
## 6 0 1 0 0
## num.of.cylindersthree num.of.cylinderstwelve fuel.system2bbl fuel.systemidi
## 1 0 0 0 0
## 2 0 0 0 0
## 3 0 0 0 0
## 4 0 0 0 0
## 5 0 0 0 0
## 6 0 0 0 0
## fuel.systemmfi fuel.systemmpfi fuel.systemspdi fuel.systemspfi
## 1 0 1 0 0
## 2 0 1 0 0
## 3 0 1 0 0
## 4 0 1 0 0
## 5 0 1 0 0
## 6 0 1 0 09.2.17 Split the data into train/test set
set.seed(123) # Setting seed for reproducibility
trainIndex <- createDataPartition(target, p = .7, list = FALSE, times = 1)
X_train <- features[trainIndex, ]
X_test <- features[-trainIndex, ]
y_train <- target[trainIndex]
y_test <- target[-trainIndex]
print(paste("Train", dim(X_train), "and test", dim(X_test)))## [1] "Train 136 and test 56" "Train 62 and test 62"9.2.18 Logarithmic scale: log base 2
alphas <- 2 ^ (2:11)
scores <- numeric(length(alphas))
for (i in seq_along(alphas)) {
lasso <- glmnet(as.matrix(X_train), y_train, alpha = 1, lambda = alphas[i])
y_pred <- predict(lasso, as.matrix(X_test), s = alphas[i], type = "response")
scores[i] <- cor(y_test, y_pred) ^ 2
}
lassocv <- cv.glmnet(as.matrix(features), target, alpha = 1, nfolds = 10)
lassocv_score <- max(lassocv$cvm)
lassocv_alpha <- lassocv$lambda.min
plot(alphas, scores, type = "b", col = "black", xlab = expression(alpha), ylab = "CV Score", log = "x")
abline(h = lassocv_score, col = "red")
print(paste("CV results:", lassocv_score, lassocv_alpha))## [1] "CV results: 65727012.0729188 43.0992704282378"10 Discussions & Conclusion
In the earlier part of this project, where we have done an extensive data cleaning & data transformation, we have identified a substantial amount of data in the features that having missing values. This creates an attention for us to handle those data first. Feature that has a number of missing value is normalized.losses and we have taken care that by using summary statistics by symboling method, which typically being used in customer segmentation.
After we have identified the categorical features, we be able to make an univariate analysis on categorical and numerical features. Analysing univariate analysis on categorical features would help us to get a glimpse of most sold cars distribution based on the features . In addition to that, univariate analysis on numerical feature has helped us in determining the skewness and detecting outliers. Based on our analysis, we have observed a right-skewed distributions were observed for normalized losses, curb weight, engine size, horsepower, and price. This indicates that while most cars fall within a typical range, there are a few outliers with significantly higher values. Features such as engine size, horsepower, curb weight, and price are key indicators of a car’s performance and market segment. This information is vital in deciding method to use in the data normalisation, which will prepared us for the next stage of modelling.
In the correlation analysis, it’s worth mentioning that we have decided to go with Pearson’s Linear Correlation Test. From this test, we can conclude that Engine Size, Curb Weight, Horsepower, Width, Length have strong positive correlations with price, indicating that they are key factors in determining the price of a car with the degree of correlation is almost at 1. While on the other hand, Normalized Losses, Stroke, Compression Ratio, Height, Symboling, Peak RPM have weak to very weak correlations with price(almost 0), indicating they have minimal influence on the pricing of cars.
In data transformation, we have created a newly defined column which is called ‘is_high_end’ to group a certain threshold of car’s price to a single category, the threshold is anything above than 75% of maximum price which is $45400, will be considered as high end. This feature is going to be the target feature for prediction model.
In preparing the data for modelling, normalization for the numerical variable and creating dummy variables for categorical data have taken place so features can better represent the underlying problem or unseen data, resulting in improved model performance. With the data set has been split into 80-20, where 80% is used for training and 20% for testing.
In our project, model evaluation has highlighted substantial improvements in performance for the classifier model, particularly after applying repeated K-fold cross-validation. Given our dataset’s relatively small size, a simple split into training and testing sets may not provide enough data for robust training and evaluation. K-fold cross-validation mitigates this by dividing the data into K parts, training the model on K-1 parts, and testing it on the remaining part. Repeating this process multiple times with different data partitions helps to produce a more reliable performance estimate, reducing variability and providing a more comprehensive evaluation of the model.
This enhanced evaluation method is crucial because our project objectives extend beyond merely predicting the correct car prices. The classification model also plays a pivotal role in marketing segmentation. By accurately classifying cars into distinct price categories, we can effectively identify and target different market segments, facilitating more precise and effective marketing strategies.
Transitioning to the regression model, our focus shifts to predicting numerical values to deepen our analysis. While the classification model aids in market segmentation, the regression model is vital for pricing rationalization. This involves modeling the relationship between the dependent variable (‘price’) and several independent variables (‘bore’, ‘stroke’, ‘horsepower’, ‘peak.rpm’, ‘curb.weight’, and ‘engine.size’) by fitting a linear equation to the dataset ‘data2’. This approach allows us to predict car prices based on their features accurately.
Moreover, we validate these predictions using Lasso regression to handle potential variations and prevent overfitting, ensuring that our model generalizes well to new data. This analysis serves dual purposes:
Pricing Rationalization: It provides a data-driven basis for setting car prices, ensuring they reflect the actual value of the vehicle. This helps car companies to stay competitive in the market while optimizing their resources. Resource Optimization: By identifying which features most significantly influence car prices, companies can prioritize their production and marketing efforts on these aspects, aligning them more closely with market demand.
In summary, integrating both classification and regression models in our analysis not only enhances our ability to predict car prices but also supports effective market segmentation. These insights are invaluable for developing targeted marketing strategies and setting rational car prices that optimize competitive positioning and resource utilization.
In essence, this project showcases how data science models can transform various facets of business operations, from enhancing customer engagement to optimizing internal processes. The insights derived from these models will empower organizations to make informed decisions, anticipate market trends, and sustain competitive advantage in a dynamic environment.