CODEWITHSUNDEEP
rggplot2
Emil Filipov
Emil Filipov

Reputation: 19

GGplot does not change color of scatter plot but does on other plots

I have a script that creates scatter a scatter plot and two bar charts from my data. Everything works fine except that when i try to change the colors of the scatter plot it does not work and they remain the standard ones.

bar chart

I want to take the red and gray colors (#f40009 and #8e8f90 respectively) from the bar chart and color the scatter plot with them, but it does not work.

Does anyone have an idea as to why this is happening and how to work it out?

The code for the scatter plot is the following:

lof <- 2
days <- 3
plot2 <-ggplot( data3, aes( CE11000_ERLOS, Z_UCS))+
  geom_point(aes(colour=ifelse( data3$LOF>lof &  data3$Z_LAST_DAYS<days &  data3$CE11000_ERLOS>100,"#8e8f90","#f40009")), size = 3)+
  labs(list(y = "Unit cases", x = "Gross sales revenue"))+
  ggtitle(bquote(atop(.("Visualization of Outliers"), atop(italic(.(country)), ""))))+
  scale_colour_discrete(guide = guide_legend(title = NULL), labels = c("Outliers", "Not outliers"))+
  theme(plot.title = element_text(hjust = 0.5))+
  geom_text(aes(label = ifelse( data3$LOF>lof &  data3$Z_LAST_DAYS<days,paste( data3$Z_CDOW, data3$CE11000_BUDAT2,sep = "\n "),""),hjust=1.05,vjust=1), size = 3.5)+
  scale_y_continuous(labels = comma)+
  scale_x_continuous(labels = comma)

print(plot2)

The data is:

LOF  Z_LAST_DAYS CE11000_ERLOS Z_UCS  Z_CDOW    CE11000_BUDAT2
3.1  1           996789        21195  Thursday  20170126
1.01 23          11912948      210839 Wednesday 20170104
1.4  22          14322767      257269 Thursday  20170105
1.01 21          11817447      185197 Friday    20170106
1.66 18          7906971       153807 Monday    20170109

And this is the result scatter

Upvotes: 0

Views: 3551

Answers (2)

user6403347
user6403347

Reputation:

Sample code to change default colors in ggplot2

data <- iris[iris$Species == "setosa" | iris$Species == "virginica", c(1:2,5)]

levels(data$Species)

droplevels(data$Species)
data

plot <- ggplot(data, aes(x = Sepal.Length, y = Sepal.Width)) + 
geom_point(aes(color = Species))

The below plot generates default colors 

plot

Plot with default colors

You can add scale_color_manual() to change the default colors

plot + scale_color_manual(
values=c("setosa" = "darkgrey","virginica" = "red"))

Plot with customized colors

Upvotes: 1

Janna Maas
Janna Maas

Reputation: 1134

in this line

geom_point(aes(colour=ifelse( data3$LOF>lof & data3$Z_LAST_DAYS<days & data3$CE11000_ERLOS>100,"#8e8f90","#f40009")), size = 3)

you're mapping the levels of what you put in the colour statement to (the default) colours by putting them inside the aes call.

If you take your colour statement out of the aes call you should get the actual colours.

compare e.g.:

not pink:

ggplot(iris, aes(Sepal.Width, Sepal.Length)) + 
  geom_point(aes(colour="pink"))

pink:

ggplot(iris, aes(Sepal.Width, Sepal.Length)) + 
  geom_point(colour="pink")

or, in your case:

geom_point(colour=ifelse( data3$LOF>lof &  data3$Z_LAST_DAYS<days &  data3$CE11000_ERLOS>100,"#8e8f90","#f40009"), size = 3)

alternatively, you can change the default colours using scale_colour_manual as per the comment by @Adam Quek.

Upvotes: 1

Related Questions