1. Nudge you towards proficiency and comfort with big data analysis
2. Develop culture of 'soups-to-nuts' reproducibility in our analytics

1. Monday: Intro to R and making simple, pretty presentations
2. Tuesday: Data wrangling
3. Wednesday: Unsupervised analysis
4. Thursday: Supervised analsyis
5. Friday: wrap up

• Morning: Background to topic, tutorials
• Break: lunch, do what you need to do
• Afternoon: Work on assignment (individually) that you will present the following morning

• Andrew Jaffe's short course: http://aejaffe.com/winterR_2016/
• R cookbook: http://www.cookbook-r.com/
• R and R packages: https://cran.r-project.org/
• Color palettes: http://colorbrewer2.org/
• R Markdown: http://rmarkdown.rstudio.com/index.html
• Embedding R code into presentations: http://zevross.com/blog/2014/11/19/creating-elegant-html-presentations-that-feature-r-code/
• Embedding R into presentations: http://ramnathv.github.io/slidifyExamples/
• Embedding R into presentations: https://www.uvm.edu/rsenr/vtcfwru/R/fledglings/14_Slideshows.html
• Git and GitHub: http://readwrite.com/2013/09/30/understanding-github-a-journey-for-beginners-part-1/
• R reference card: http://cran.r-project.org/doc/contrib/Short-refcard.pdf
• Online interactive exercise: http://tryr.codeschool.com/
• Free and excellent text editor with R syntax highlighting: http://www.sublimetext.com/2

1. Evernote for writing code, then cut and paste into terminal.
2. Text editor (Sublime 2) for writing functions

In slides, a command (we'll also call them code or a code chunk) will look like this

print("I'm code")

## [1] "I'm code"

And then directly after it, will be the output of the code. So print("I'm code") is the code chunk and [1] "I'm code" is the output.

2 + (2 * 3)^2

## [1] 38

(1 + 3) / 2 + 45

## [1] 47

1:10

##  [1]  1  2  3  4  5  6  7  8  9 10

1:10 is equivalent to the seq function

seq(from=1,to=10,by=1)

##  [1]  1  2  3  4  5  6  7  8  9 10

To learn more about a function, use help

?seq

Most of the time you want to capture the results of a computation. Variables

1. Create variables from within the R environment and from files on your computer
2. Use "=" or "<-" to assign values to a variable name
3. Case-sensitive

x <- rnorm(1e4, mean=0, sd=2) # result does not get sent to output

R's listing function, ls:

ls()

## [1] "x"

Things can be done with variables

mean(x);

## [1] 0.03374283

hist(x);

• Classes define the structure of a variable, and the set of operations or functions that can performed on them
• vector and matrix are classes
• data.frame is one of the most frequently used classes of R variables
• data.frame is like a spreadsheet. Rows are entities and columns are variables characterizing that entity.

• data.frames are somewhat advanced objects in R

• Here we introduce "1 dimensional" classes; these are often referred to as 'vectors'

• Vectors can have multiple sets of observations, but each observation has to be the same class

class(x)

## [1] "numeric"

y = "mistakenly, embryomics, smoogy boogy"
print(y)

## [1] "mistakenly, embryomics, smoogy boogy"

class(y)

## [1] "character"

The function c() collects/combines/joins single R objects into a vector of R objects. It is mostly used for creating vectors of numbers, character strings, and other data types.

x <- c(1, 4, 6, 8)
x

## [1] 1 4 6 8

class(x)

## [1] "numeric"

length(): Get or set the length of vectors (including lists) and factors, and of any other R object for which a method has been defined.

length(x)

## [1] 4

y

## [1] "mistakenly, embryomics, smoogy boogy"

length(y)

## [1] 1

Go to http://tryr.codeschool.com/ and earn your chapter badges!

Load the data, modify path accordingly...

survdata<-read.csv("../../resources/character-deaths.csv", as.is=TRUE)
dim(survdata)

## [1] 917  13

colnames(survdata)

##  [1] "Name"               "Allegiances"        "Death.Year"        
##  [4] "Book.of.Death"      "Death.Chapter"      "Book.Intro.Chapter"
##  [7] "Gender"             "Nobility"           "GoT"               
## [10] "CoK"                "SoS"                "FfC"               
## [13] "DwD"

Use ggplot2 to make nice graphics

Install the package if you don't already have it...

install.packages("ggplot2")

Load the library

library(ggplot2)

How many characters per house?

ggplot(survdata, aes(x="Allegiances")) + geom_bar(stat="bin") + theme_bw()

How many characters per house?

ggplot(survdata, aes(x=Allegiances)) + geom_bar(stat="bin") + theme_bw()

How many characters per house?

ggplot(survdata, aes(x=Allegiances)) + geom_bar(stat="bin") + theme_bw() + theme(text = element_text(size=8), axis.text.x = element_text(angle=90, vjust=0.5, hjust=1)) +
    theme(axis.title.x = element_blank())+ ylab("") + xlab("") + coord_flip()

length(which(is.na(survdata$Death.Year)))

## [1] 612

length(which(!is.na(survdata$Death.Year)))

## [1] 305

isDead<-rep("Dead", nrow(survdata))
isDead[ which( is.na(survdata$Death.Year)) ] <-"Alive"
survdata<-cbind(survdata, isDead=isDead)
ggplot(survdata, aes(x=Allegiances, fill=as.factor(isDead))) + geom_bar(stat="bin") + theme_bw() + theme(text = element_text(size=8), axis.text.x = element_text(angle=90, vjust=0.5, hjust=1)) +
    theme(axis.title.x = element_blank())+ ylab("") + xlab("") + coord_flip()

ggplot(survdata, aes(x=Death.Chapter)) + geom_histogram(colour="black", fill="white") + facet_grid(. ~ Book.of.Death) + theme_bw()

## stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.
## stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.
## stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.
## stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.
## stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.
## stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.

Download from http://www.cahanlab.org/intra/training/bootcampJune2016/misc/timesData.csv

1. Use facets to plot distributions of teaching scores, one plot per year
2. Scatter plot to explore relationship between the teaching score and the income (facet by year)
3. Add a regression line to the above plot
4. Same as 2 and 3 but exploring the relationship between total_score and income
5. Same as 2 and 3 but exploring relationship between world_rank and research
6. Does there appear to be an effect of female_male_ratio and total_score? Use boxplot to visualize