R coloring and adding lines on a loop

Question

I want to generate a ggplot by adding different lines on a loop. Each line should have its own color. I have attempted this:

scenario_colors <- c(
  "scenario1" = "black",
  "scenario2" = "red",
  "scenario3" = "green"
)
all_scenario_names <- c("scenario1", "scenario2", "scenario3")
p <- p + scale_color_manual(values = scenario_colors)

for (j in 1:3){
  column <- column_names[j]
  scenario <- all_scenario_names[j]
  print(paste0(column, " ---- ", scenario, " ---- ", scenario_colors[scenario]))
  p <- p + geom_line(data=rama_data, aes(y = .data[[column]], group = 1, color=scenario))
}

This however causes an erroneous final plot. All 3 lines are the color of scenario 3.

But if i try to hard-code it it works fine

for (j in 1:3){
  column <- column_names[j]
  scenario <- all_scenario_names[j]
  print(paste0(column, " ---- ", scenario, " ---- ", scenario_colors[scenario]))
  if (j == 1){p <- p + geom_line(data=rama_data, aes(y = .data[[column]], group = 1, color="scenario1"))}
  else if(j == 2){p <- p + geom_line(data=rama_data, aes(y = .data[[column]], group = 1, color="scenario2"))}
  else{p <- p + geom_line(data=rama_data, aes(y = .data[[column]], group = 1, color="scenario3"))}
}

Each line is shown in its corresponding color (black, red and green).

I have been stuck on this piece of code for a couple of days now and can't seem to figure out what is wrong with the first snippet. This second approach is unfeasible since in the future i will need to expand the amount of scenarios. Hardcoding them by hand would not be possible. What is wrong in the first approach?

Extra information: The shape of the data used is like this:

variable1 <- c(runif(5))
variable2 <- c(3*runif(5))
variable3 <- c(5 + runif(5))
period <- c("MM012021", "MM022021", "MM032021", "MM042021", "MM052021")
# Create the data.table
column_names <- c("variable1", "variable2", "variable3")
rama_data <- data.table(period, variable1, variable2, variable3)

I also add the results between both implementations: Thanks!

Answer 1

You can handle this in one ggplot2::geom_line() call with any number of scenarios by mapping the color aesthetic to each scenario. To do this, ggplot2 expects the data in long form. As you're using data.table, let's do it that way.

rama_long <- melt(
    rama_data,
    id.vars = "period",
    variable.name = "scenario"
)[
    ,
    scenario := sub("variable", "", scenario)
][
    ,
    period := as.IDate(
        period,
        format = "MM%m%d%y"
    )
]

Note, I also made your period into a date. Then we can just supply the scenario column to the color aesthetic:

ggplot(rama_long) +
    geom_line(
        aes(
            x = period,
            y = value,
            color = scenario
        ),
        linewidth = 2
    )