INTRODUCTION
Exploring Data
The dataset about car price prediction was downloaded from kaggle.
It was then read into R cloud and further the following analysis were done - subsetting, outliers detection, Grammar of graphics plots such as histogram,boxplot and scatter plot analysis.
The initial assumptions are as follows:
Higher the horsepower, more the number of cylinders
Higher the horsepower, lower the mpg
There exists a strong correlation between horsepower and price
Fuel type determines the mileage to some extent
======================================================================
======================================================================
CORRELATION
Correlation factors
[1] -0.8014562
[1] -0.7705439
[1] 0.8081388
[1] -0.6857513
[1] -0.6975991
HISTOGRAM
HORSEPOWER
FUEL TYPE
CYLINDER NUMBER
BOXPLOT
# Analysis
HORSE POWER AND PRICE
PRICE AND CYLINDER NUMBER
HORSE POWER AND CYLINDER
HORSEPOWER AND DRIVEWHEEL
HORSEPOWER AND FUELTYPE
# PLOTS {.tabset}
INSIGHTS
Inferences observed
There exists a strong correlation between hosrepower and price of the car.
cor=0.8014562– Thus both the factors are highly influencing factors. Higher the horsepower, higher the price of the vehicles.
Horsepower in front wheel drive(fwd) is maximum.
Diesel vehicles exhibit slightly greater city and highway mileage than gas vehicles.
Average and idle number of cylinders lies in the range of 4-6.
Mileage in city as well as highways reduces with increase in horsepower.
The ideal price of the car with good mileage and horsepower can be between 20000 to 30000.
---
title: "ANALYSIS ON ATTRIBUTES WHICH INFLUENCES CARS"
output:
flexdashboard::flex_dashboard:
orientation: columns
vertical_layout: fill
theme: darkly
storyboard: yes
social: menu
source_code: embed
html_document:
df_print: paged
---
```{r setup, include=FALSE}
libraries= c("flexdashboard","lattice","DT","ggplot2")
lapply(libraries,require,character.only=TRUE)
data=read.csv(file.choose())
new_data<-data[c("CarName","fueltype","drivewheel","cylindernumber","horsepower","citympg","highwaympg","price")]
attach(data)
```
# INTRODUCTION {.tabset}
### Exploring Data
### The dataset about car price prediction was downloaded from kaggle.
### It was then read into R cloud and further the following analysis were done - subsetting, outliers detection, Grammar of graphics plots such as histogram,boxplot and scatter plot analysis.
### The initial assumptions are as follows:
### Higher the horsepower, more the number of cylinders
### Higher the horsepower, lower the mpg
### There exists a strong correlation between horsepower and price
### Fuel type determines the mileage to some extent
======================================================================
======================================================================
# MENU {.tabset}
### Structure
```{r}
str(data)
```
### SUMMARY
```{r}
summary(data)
```
# CORRELATION {.tabset}
### Correlation factors
```{r}
cor(horsepower,citympg)
cor(horsepower,highwaympg)
cor(horsepower,price)
cor(citympg,price)
cor(highwaympg,price)
```
# HISTOGRAM {.tabset}
### Comparison of attributes
```{r}
hist(new_data$price,main="PRICE FREQUENCY")
hist(new_data$citympg,main="MILEAGE IN CITY ROADS")
hist(new_data$highwaympg,main="MILEAGE IN HIGHWAYS")
hist(new_data$horsepower,main="HORSEPOWER OF THE CARS")
```
-----------------------------------------------------------------------
### HORSEPOWER
```{r}
histogram(~horsepower|drivewheel,data=new_data)
histogram(~horsepower|cylindernumber,main="HORSEPOWER VS CYLINDERS")
```
### FUEL TYPE
```{r}
histogram(~citympg|fueltype,main="CITY MPG VS FUEL")
histogram(~highwaympg|fueltype,main="HIGHWAY MPG VS FUEL")
```
### CYLINDER NUMBER
```{r}
histogram(~price|cylindernumber,main="CYLINDERS VS PRICE")
histogram(~horsepower|cylindernumber,main="HORSEPOWER VS CYLINDERS")
```
# BOXPLOT {.tabset}
# Analysis {.tabset}
-------------------------------------------------------------------
------------------------------
### HORSE POWER AND PRICE
```{r}
ggplot(new_data)+geom_boxplot(aes(x=horsepower,y=price,fill=horsepower))+ggtitle('BOXPLOT OF HORSEPOWER AND PRICE')+xlab('HORSEPOWER')+ylab('PRICE')
```
### PRICE AND CYLINDER NUMBER
```{r}
ggplot(new_data)+geom_boxplot(aes(x=price,y=cylindernumber,fill=price))+ggtitle('BOXPLOT OF PRICE AND CYLINDER NUMBERS')+xlab('PRICE')+ylab('CYLINDERNUMBERS')
```
### HORSE POWER AND CYLINDER
```{r}
ggplot(new_data)+geom_boxplot(aes(x=horsepower,y=cylindernumber,fill=horsepower))+ggtitle('BOXPLOT OF HORSEPOWER AND CYLINDERS')
```
### HORSEPOWER AND DRIVEWHEEL
```{r}
ggplot(new_data)+geom_boxplot(aes(x=horsepower,y=drivewheel,fill=horsepower))+ggtitle('BOXPLOT OF HORSEPOWER AND DRIVEWHEEL')
```
### HORSEPOWER AND FUELTYPE
```{r}
ggplot(new_data)+geom_boxplot(aes(x=horsepower,y=fueltype,fill=horsepower))+ggtitle('BOXPLOT OF HORSEPOWER AND FUELTYPE')
```
# PLOTS {.tabset}
### Simple Plots
```{r}
ggplot(new_data)+geom_point(aes(x=horsepower,y=price))
ggplot(new_data)+geom_point(aes(x=price,y=cylindernumber))
ggplot(new_data)+geom_point(aes(x=horsepower,y=cylindernumber))
ggplot(new_data)+geom_point(aes(x=horsepower,y=citympg))
ggplot(new_data)+geom_point(aes(x=horsepower,y=highwaympg))
ggplot(new_data)+geom_point(aes(x=horsepower,y=drivewheel))
ggplot(new_data)+geom_bar(aes(x=horsepower,stat="bin",binwidth=0.05))
```
# DATATABLE {.tabset}
### Valuebox
```{r}
value=sum(car_ID)
valueBox("count of cars",value,icon="fa fa-user")
```
### DOWNLOAD
```{r}
datatable(data,extensions='Buttons',options=list(dom="Bftrip",Buttons=c('copy','print','csv','pdf')))
```
# INSIGHTS {.tabset}
### Inferences observed
### There exists a strong correlation between hosrepower and price of the car.
### cor=0.8014562-- Thus both the factors are highly influencing factors. Higher the horsepower, higher the price of the vehicles.
### Horsepower in front wheel drive(fwd) is maximum.
### Diesel vehicles exhibit slightly greater city and highway mileage than gas vehicles.
### Average and idle number of cylinders lies in the range of 4-6.
### Mileage in city as well as highways reduces with increase in horsepower.
### The ideal price of the car with good mileage and horsepower can be between 20000 to 30000.
