Plotting_Examples.Rmd

---
title: "Plotting examples in R base graphics and ggplot"
author: "Katie Gostic"
date: "February 23, 2016"
output: pdf_document
---

```{r, echo = FALSE}
require(knitr)
```

### Load packages and data

```{r}
#Load the ggplot package
library(ggplot2)

data(mpg) #Load built-in data set, mpg
?cars #Inspect your data set and get to know the variables
head(mpg)


```

### Section 1. Examples in R base graphics using the mpg dataset

```{r}

attach(mpg) #Attach the data frame so you can call columns by name
              # e.g., other wise you would need to type mpg$hwy instead of just hwy


plot(cty, hwy, xlab = 'Highway Mileage', ylab = 'City Mileage', main = 'MPG Plot')

# Jitter the points so they don't all overlap
plot(jitter(cty), jitter(hwy), xlab = 'Highway Mileage', ylab = 'City Mileage', 
     main = 'MPG Plot')

#Change the color
plot(jitter(cty), jitter(hwy), xlab = 'Highway Mileage', ylab = 'City Mileage', 
     main = 'MPG Plot', col = 'steelblue2')

#Change the point type
plot(jitter(cty), jitter(hwy), xlab = 'Highway Mileage', ylab = 'City Mileage',
     main = 'MPG Plot', col = 'steelblue2', pch = 19)
# See ?points for a list of point characters

#Change the point size
plot(jitter(cty), jitter(hwy), xlab = 'Highway Mileage', ylab = 'City Mileage', 
     main = 'MPG Plot', col = 'steelblue2', pch = 19, cex = 2)

# Change the text size of the title, axis labels (words) and axis ticks (numbers on tick marks)
#Change the point type
plot(jitter(cty), jitter(hwy), xlab = 'Highway Mileage', ylab = 'City Mileage', 
     main = 'MPG Plot', col = 'steelblue2', pch = 19, cex.main = 2) #title size

plot(jitter(cty), jitter(hwy), xlab = 'Highway Mileage', ylab = 'City Mileage', 
     main = 'MPG Plot', col = 'steelblue2', pch = 19, cex.lab = 2) #Axis label

plot(jitter(cty), jitter(hwy), xlab = 'Highway Mileage', ylab = 'City Mileage', 
     main = 'MPG Plot', col = 'steelblue2', pch = 19, cex.axis = 2) #Axis tick


# Keep your memory clear of clutter by detaching data frames when 
#you are done plotting
detach(mpg)


# Plot Audi and Chevrolet data in different colors

#First, take subsets of the full data set
Audi = subset(mpg, subset = manufacturer == 'audi')
Chevrolet = subset(mpg, subset = manufacturer == 'chevrolet')

#Start by plotting the Audi data
plot(jitter(Audi$cty), jitter(Audi$hwy), xlab = 'Highway Mileage', ylab = 'City Mileage', main = 'MPG Plot', col = 'steelblue2', pch = 19)

#Overlay the chevy plot using "points" or "lines"
points(jitter(Chevrolet$cty), jitter(Chevrolet$hwy), pch = 19, col = 'magenta')

#Add a legend in the top left
legend('topleft', legend =  c('Audi', 'Chevrolet'), fill = c('steelblue2', 'magenta'))
?legend

```

### Section 2. The same examples, plotted using ggplot2
##### NOTE: you MUST put your data into a data frame to use ggplot2

```{r}
# Make a basic plot
p = ggplot(mpg, aes(x = cty, y = hwy))
```

In ggplot syntax, you start with the name of your data frame, here mpg, and use the aesthetics option to tell it which columns to use on the x and y axis. Save these instructions to variable p. Then plot the basic plot layer, p below, using the + sign to add new plot layers (below layers include plot geometry, titles, etc.)

```{r}
p + geom_point() #Show plot with basic geometry

p + geom_point(color = 'steelblue2') #Add color

p + geom_jitter(color = 'steelblue2') #Add jitter

p + geom_jitter(color = 'steelblue2', size = 3) #Change point size

p + geom_jitter(color = 'steelblue2', size = 3) + ggtitle('MPG plot') #Add title

p + geom_jitter(color = 'steelblue2', size = 3) + ggtitle('MPG plot') + 
  theme(axis.title = element_text(size=22)) +                              
  #Change axis label size
  theme(plot.title = element_text(size=22))                                 
#Change plot title size

p + geom_jitter(aes(color = model)) # Determine point color based on 
#the values in the "model" column

p + geom_jitter(aes(color = manufacturer)) # Determine point color based 
#on the values in the "manufacturer" column

p + geom_jitter(color = 'steelblue2') + geom_smooth(method = lm) # Add a 
#regression line with confidence intervals automatically plotted

p + geom_jitter(color = 'steelblue2') + geom_smooth(method = lm, se = FALSE) 
# Add a regression line without confidence intervals
# Note, you use similar modifications to change the point size

## See the ggplot documentation for help, or Google for example 
#scripts that make the plots you want
```

This page was made using R Markdown with package knitr.