Introductory Stata 43: Graphs For Two Continuous Variables (Line plots)

Let's continue to examine the graphs for two continuous variables. Today, we will take a look at the line plots. Line plots are often used to show the trend of a variable over time. The time variable is often placed on the x-axis. If the x variable is not a time variable in natural order, we should have the data sorted before we draw the line plots.

Let's open a dataset I created from the Current Population Survey (CPS) for the households in the United States.

********************
*43. Line plots *
********************
capture log close

set showbaselevels on

*Use the dataset I created for US households

*line plot
graph twoway (line income year)

graph twoway (line income year, xlabel(2000(1)2021, angle(45)))

*line options
graph twoway (line income year, connect(stairstep))
graph twoway (line income year, connect(stepstair))

graph twoway (line income year, lpattern(dash) lcolor(blue) lwidth(thick))

*line plot with scatterplot
graph twoway (line income year) (scatter income year)
graph twoway (scatter income year, connect(l))

*multiple line plots
graph twoway (line inc_metro year) (line inc_nonmetro year)

graph twoway (line inc_metro year, lcolor(blue)) (line inc_nonmetro year, lcolor(orange)), legend(order(1 "Household income in metro area" 2 "Household income in non-metro area") col(1))

*multiple line plots with two y axes
generate difference=inc_metro-inc_nonmetro

graph twoway (line inc_metro year) (line inc_nonmetro year) (line difference year), legend(order(1 "Household income in metro area" 2 "Household income in non-metro area" 3 "Difference") col(1))

graph twoway (line inc_metro year, lcolor(blue) yaxis(1)) (line inc_nonmetro year, lcolor(orange) yaxis(1)) (line difference year, lpattern(dash) lcolor(black) yaxis(2)), legend(order(1 "Household income in metro area" 2 "Household income in non-metro area" 3 "Difference") col(1))

log close

#Line #TwoYAxes