describing data : continuous variables

[woneuy01]

describing data : continuous variables

str(cars) mean(cars$mpq) median(cars$cyl) varation <- sd(cars$mpq) range(cars$mpq)

quantile(cars$mpg) 0% 25% 50% 75% 100% 10.400 15.425 19.200 22.800 33.900

quantile(cars$mpg, probs = c(0.05, 0.95))
5% 95% 11.995 31.300

[woneuy01]

Describing Categories Counting appearance Creating a table

amtable <- table(cars$am) amtable
auto manual
13 19 amtable / sum(amtable) auto manual 0.40625 0.59375 sapply(mtcars, function(x) length(unique(x))) # low numbers can convert to factors

[woneuy01]

Describing Distributions plot histogram

hist(cars$mpg, col = "grey") R and you want to have bars representing the intervals 5 to 15, 15 to 25, and 25 to 35, hist(cars$mpg, breaks = c(5, 15, 25, 35))

[woneuy01]

By breaking up your data in intervals, you still lose some information, Still, the most complete way of describing your data is by estimating the probability density function (PDF) or density of your variable. mpgdens <- density(cars$mpg) plot(mpgdens)

Plotting densities in a histogram

hist(cars$mpg, col = "grey", freq = FALSE) lines(mpgdens)

[woneuy01]

Describing Multiple Variables Summarizing a complete dataset

summary(cars)
mpg cyl am gear
Min. :10.40 Min. :4.000 1st Qu.:15.43 1st Qu.:4.000 manual:19 4:12 Median

[woneuy01]

Plotting quantiles for subgroups One way to quickly compare groups is to construct a box‐and‐whisker plot from the data "plot boxes for the variable mpg for the groups defined by the variale cyl"

boxplot(mpg ~ cyl, data = cars)

[woneuy01]

Correlations plot(iris[-5]) pairs() #create plot matrix

with(iris, cor(Petal.Width, Patal.Length)) [1]0.9628654 iris.cor <- cor(iris[-5]) str(iris.cor) iris.cor["Patal.Width","Petal.Length"] [1] 0.9628654

[woneuy01]

sort table sort(table(x))