Вопрос

So I used the following code to generate a scatter plot with a standard error

data <- ddply(dat, .(Value), summarise, 
               N    = length(means),
               mean = mean(means),
               sd   = sd(means),
               se   = sd(means) / sqrt(length(means)) )
ggplot(data, aes(x=Value, y=mean)) + 
    geom_errorbar(aes(ymin=mean-se, ymax=mean+se), width=.1) +
    geom_line() +
    geom_point()

Here is sample data

Value  N     mean        sd         se
     1 11 1.624771 0.1788739 0.05393250
     2  6 1.775057 0.2625611 0.10719012
     3 11 2.218854 0.4320835 0.13027807
     4 10 1.745128 0.3922374 0.12403637
     5  9 2.266107 0.1645616 0.05485388

So what I want to try to do is find the 'area' that the standard errors occupy and give a graph that has the standard error areas occupied. Is this possible ? So I just basically want the area occupied by the max standard error and a min standard error for each data point

Это было полезно?

Решение

Use geom_ribbon:

ggplot(data, aes(x=Value, y=mean)) + 
  geom_errorbar(aes(ymin=mean-se, ymax=mean+se), width=.1) +
  geom_ribbon(aes(ymin=mean-se, ymax=mean+se),alpha=0.5) +
  geom_line() +
  geom_point()

enter image description here

Лицензировано под: CC-BY-SA с атрибуция
Не связан с StackOverflow
scroll top