Introduction

Questions

Variation

Visualizing distributions

diamonds %>%
    ggplot(aes(x = cut)) + geom_bar()

diamonds %>% ggplot(mapping = aes (x = carat)) +
    geom_histogram(binwidth = 0.5)

diamonds %>%
    filter(carat < 3) %>%
    ggplot(aes(x = carat)) + geom_histogram(binwidth = 0.5)

diamonds %>%
    ggplot(aes(x = carat, color = cut)) + geom_freqpoly()
## `stat_bin()` using `bins = 30`. Pick better value `binwidth`.

Typical values

diamonds %>%
    
    # Filter out diamonds > 3 carat
    filter(carat > 3) %>%
    
    # Plot
    ggplot(aes(x = carat)) + geom_histogram(binwidth = 0.01)

faithful %>% ggplot(aes(eruptions)) + geom_histogram(binwidth = 0.25)

Unusual values

diamonds %>%
    ggplot(aes(y)) + geom_histogram()
## `stat_bin()` using `bins = 30`. Pick better value `binwidth`.

diamonds %>%
    ggplot(aes(y)) + geom_histogram() + coord_cartesian(ylim = c(0, 50))
## `stat_bin()` using `bins = 30`. Pick better value `binwidth`.

Missing Values

diamonds %>%
    # filter(y < 3 | y > 20) %>%
   
     mutate(y = ifelse(y < 3 | y > 20, NA, y)) %>%
    
    # Plot
    ggplot(aes(x = x, y = y)) + geom_point()
## Warning: Removed 9 rows containing missing values or values outside the scale range
## (`geom_point()`).

Covariation

A Categorical and continuous variable

diamonds %>% 
    ggplot(aes(x = cut, y = price)) + geom_boxplot()

Two categorical variables

diamonds %>%
    count(color, cut) %>%
    ggplot(aes(x = color, y = cut, fill = n)) + geom_tile()

Two continous variables

library(hexbin)
diamonds %>%
    filter(carat < 3) %>%
    ggplot(aes(x = carat, y = price)) + geom_hex()

diamonds %>%
    ggplot(aes(x = carat, y = price)) + geom_boxplot(aes(group = cut_width(carat, 0.1)))

## Patterns and models

library(modelr)
mod <-lm(log(price) ~ log(carat), data = diamonds)

diamonds4 <- diamonds %>%
    modelr::add_residuals(mod) %>%
    mutate(resid = exp(resid))

diamonds4 %>%
    ggplot(aes(carat, resid)) + geom_point()

diamonds4 %>% 
    ggplot(aes(cut,resid)) + geom_boxplot()