plots

ggplot2

• library(ggplot2)

• ggplot(df, aes(x=col1, y=col2)) +

  • geom_point() + # scatter plot

  • geom_line() + # line

  • geom_smooth() + # regression/smoother line

  • geom_histogram() + # histogram (use aes(x=col) only)

  • labs(x="X label", y="Y label", title="Title") +

  • theme(plot.title=element_text(hjust=0.5)) # center title

• aes() maps data columns to visual properties. Add layers with +.

qplot

• qplot(x, y) # scatter

• qplot(x, y, geom="line") # line

• qplot(values, geom="histogram") # histogram

• Shortcut for simple plots. Less flexible than full ggplot but quick for exploration.

to customize

• # color by group

  • aes(x=col1, y=col2, color=group_col)

• # size of points

  • geom_point(size=2)

• x axis as dates — works automatically if col is Date type

• # label specific points

  • geom_smooth(method="lm") # linear regression line specifically

plotly

• library(plotly)

• plot_ly(data=df, x=~col1, y=~col2,

• type="scatter", mode="markers")

• Note the ~ before column names. Makes charts zoomable and hoverable.

using pipes in plots

df %>%

filter(col > 5) %>%

ggplot(aes(x=col1, y=col2)) +

geom_point()

• Use data=. when piping — the dot means "use whatever came through the pipe."