November 14th, 2016
Telling stories
Descriptive Statistics
We have data, we want to tell something about them
We use some functions to analyze our data
- Counting them
- Locating them
- Describe them
Functions
Have a name, several inputs and one output
Inputs are always inside round parenthesis ( )
Some inputs can be optional. They have default values
You have to read the manual using
help(function_name)
or in short version
?function_name
Functions used to describe vectors
There are many. You have to explore and learn
length()min(),max(),range()head(),tail()summary()table()
Describing factor vectors
length(state.region)
[1] 50
summary(state.region)
Northeast South North Central West
9 16 12 13
table(state.region)
state.region
Northeast South North Central West
9 16 12 13
Describing character vectors
length(state.abb)
[1] 50
summary(state.abb)
Length Class Mode
50 character character
table(state.abb)
state.abb AK AL AR AZ CA CO CT DE FL GA HI IA ID IL IN KS KY LA MA MD ME MI MN MO MS 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 MT NC ND NE NH NJ NM NV NY OH OK OR PA RI SC SD TN TX UT VA VT WA WI WV WY 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
Describing numeric vectors
length(state.area)
[1] 50
summary(state.area)
Min. 1st Qu. Median Mean 3rd Qu. Max. 1214 37317 56222 72368 83234 589757
table(state.area)
state.area
1214 2057 5009 6450 7836 8257 9304 9609 10577 24181
1 1 1 1 1 1 1 1 1 1
31055 33215 36291 40395 40815 41222 42244 45333 47716 48523
1 1 1 1 1 1 1 1 1 1
49576 51609 52586 53104 56154 56290 56400 58216 58560 58876
1 1 1 1 1 1 1 1 1 1
68192 69686 69919 70665 77047 77227 82264 83557 84068 84916
1 1 1 1 1 1 1 1 1 1
96981 97914 104247 110540 113909 121666 147138 158693 267339 589757
1 1 1 1 1 1 1 1 1 1
Data Visualization
“one image worths a thousand words”
Graphics
Sometimes the best way to tell the story of the data is with a graphic
plot(state.area)
Another
plot(rivers)
Each element has a position in the x axis
Plotting Factors
The previous graphics used numeric data. What about factors?
plot(state.region)
Barplots
- Numeric vectors are shown element by element
- Factor vectors are shown as a “table”
- i.e. the frequency of each value
- i.e. counting how many times for each value
- Can we do the same for a numeric vector?
- all values are different
- we have to group them in “similar” sets
Histograms
plot(rivers)
hist(rivers)
Histograms
Numeric data is grouped in N classes
hist(rivers, col="grey")
hist(rivers, col="grey", nclass = 30)
Making it beautiful
Choosing the color
plot(rivers)
plot(rivers, col="red")
Choosing the size of the symbol
plot(rivers, cex=2)
plot(rivers, cex=0.5)
Choosing the symbol
plot(rivers, pch=16)
plot(rivers, pch=".")
Choosing the type of plot
plot(rivers, type = "l")
plot(rivers, type = "o")
Zooming
plot(rivers, type = "l", xlim=c(1,50))
plot(rivers, type = "o", xlim=c(100,141))
More plot types
plot(rivers, type = "b")
plot(rivers, type = "n")
Full annotation
plot(rivers, main = "Length of Rivers", sub = "141 samples", ylab="Length [miles]")
Two or more variables
Two plots in parallel
plot(state.x77[,"Area"], ylab="Area [sq mi]") points(state.x77[,"Population"]*10, pch=2)
The first one defines the scale
Two lines in parallel
plot(state.x77[,"Area"], type="l", ylab="Area [sq mi]") lines(state.x77[,"Population"]*10, col="red")
