R图中的图例显示所有线条

Question

I have a following problem. I want to put a legend into my graph. My code:

plot(Lc(`BEL_2016_final.csv`$value),col="red",lwd=2, 
     xaxt="n", yaxt="n", cex.lab = 1.5)
axis(side=1, at=axTicks(1), cex.axis = 1.5)
axis(side=2, at=axTicks(2), cex.axis = 1.5)

par(new=TRUE)

plot(Lc(`CRO_2016_final.csv`$value),col="blue",lwd=2,
     xaxt="n", yaxt="n", cex.lab = 1.5)
axis(side=1, at=axTicks(1), cex.axis = 1.5)
axis(side=2, at=axTicks(2), cex.axis = 1.5)

legend(x = "topleft", legend=paste0(c("Belgium, Gini "),
                            round(Gini(`BEL_2016_final.csv`$value), digits = 2),
                            c("Croatia, Gini "), 
                            round(Gini(`CRO_2016_final.csv`$value), digits = 2)),
       col=c("red", "blue"), lty=1:2, cex=1, lwd=1.5)

However, the legend looks like this:

When I try:

legend=paste0(c("Belgium, Gini ", "Croatia, Gini "),
                                    round(c(Gini(`BEL_2016_final.csv`$value)),
                                    Gini(`CRO_2016_final.csv`$value)),
                                   digits = 2)

I got this result:

which is wrong, because Gini index for Croatia is 0.73.

How can I modify my code to display both lines (red and blue) in the legend, both on a new line? Thanks a lot.

Answer 1

Your parentheses are mismatched. Whatever IDE/editor you are using I encourage the use of matching (sometimes "rainbow") parentheses. For example, in RStudio, if the cursor is the _ symbol (and accepting RStudio's insistence on its indentation preference):

notice that the ( next to paste0 is highlighted, suggesting you that digits=2 is the last argument in paste0 . This is incorrect. Another hint is using RStudio's indentation preference (highlight the block and press Ctrl-I , the default keypress for "Reindent Lines"): the second Gini lines up with c( , not with the first Gini , meaning that c( and second- Gini are at the same level ... where I would expect the second- Gini to be nested within the c( .

To validate what is going on, I'll replace the Gini(.) calls with your 0.52 and 0.73 values, verbatim (but please keep them as Gini(.) in your code:

paste0(c("Belgium, Gini ", "Croatia, Gini "),
                                    round(c(0.52),
                                    0.73),
                                   digits = 2)
# [1] "Belgium, Gini 0.52" "Croatia, Gini 0.52"

Looking at it this way, it appears as if the first right-paren after 0.52 might have been intended to be after the 0.73 , since grouping 0.52 and 0.73 makes sense.

Here is corrected code, where all I do is remove one right-paren from after the first- Gini , and add one right-paren to the very end of this expression:

legend=paste0(c("Belgium, Gini ", "Croatia, Gini "),
              round(c(Gini(`BEL_2016_final.csv`$value),
                      Gini(`CRO_2016_final.csv`$value)),
              digits = 2) )

and the associated matching-paren highlighting (again, _ is the current cursor):

<soapbox>

PS: I am not saying that one must use the RStudio IDE for R work. In fact, I don't, I use emacs/ess. There are other editors to use as well. However, as much as indentation and similar can be viewed as style and therefore not important for programming, I argue that indentation and some editor functionality like matching-parens can help in readability as well as troubleshooting code before you even get to a mistake; for instance, a consistent indentation style alone here hints to improper paren-closure, and the matching-paren-highlighter confirms it. Use what you prefer, but some programming styles are actually beneficial functionally (and therefore pragmatic).

</soapbox>

Answer 2

Correct solution is:

legend(x = "topleft", legend=paste0(c("Belgium, Gini ", "Croatia, Gini "),
                                    c(round(Gini(`BEL_2016_final.csv`$value), digits = 2),
                                    round(Gini(`CRO_2016_final.csv`$value), digits = 2)
                                    )),
       col=c("red", "blue"), lty=1:2, cex=1, lwd=1.5)