how to normalize a barplot to normal distribution in R?

Question

I created a barplot using ggplot in which the x-axis represents a number in acending order and each number has is own percentage. I would like to normalized the observed barplot (Gaussian) in order to compare the two barplots. someone know how to do it? Here is my code:

lemon_imp_05 = subset(Lemon_brevante_data,Lemon_brevante_data$Block == "IMP-05")
  S = lemon_imp_05$percentage
  names(S) = lemon_imp_05$count
  barplot(S, main = 'Block IMP-05 Loam',
          xlab= 'Count(N.fruits/carton)',ylab = 'percentage(%)', col = "green")

The result is a non normal distribution bar plot, but I want to force a normal distribution on the observed barplot (don't know how to add picture of the current result). any suggestions? Thanks

Normalize your data. barplot(S / sum(S), ...). Though this would be a sum-to-one normalization, not a Gaussian normalization. You could use scale(S) instead if you want to center and scale the vector. If you edit your question, in the edit bar there's an "add image" button. — Gregor Thomas
Do I need to write: barplot(scale(S),...) ? and in barplot(S / sum(S),...) I can't see any difference. or maybe I don't write it in the right position — Erez Kauffman
Could you share some sample data? Something like dput(S[1:10]) for the first 10 values would be great - it will be copy/pasteable and include all structure and class information. — Gregor Thomas

jay.sf jay.sf · Accepted Answer · 2021-01-13T19:50:07

Do you want to plot the densitity distribution of your empirical data, and compare it to a normal distribution with same mean and sd?

plot(density(y), ylim=c(0, .5), lwd=2)
curve(dnorm(x, mean=mean(y), sd=sd(y)), add=TRUE, lwd=2, lty=2, col=2)

Like this?

Data:

set.seed(42)
y <- rnorm(1e2, mean=1.5, sd=.8)

Or use hist.

h <- hist(y, freq=F, breaks="FD")
curve(dnorm(x, mean=mean(y), sd=sd(y)), add=TRUE, lwd=2, lty=2, col=2)

## stored values
h
# $breaks
# [1] -1.0 -0.5  0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5
# 
# $counts
# [1]  2  2  6 15 21 23 17 11  3
# 
# $density
# [1] 0.04 0.04 0.12 0.30 0.42 0.46 0.34 0.22 0.06
# 
# $mids
# [1] -0.75 -0.25  0.25  0.75  1.25  1.75  2.25  2.75  3.25
# 
# $xname
# [1] "y"
# 
# $equidist
# [1] TRUE
# 
# attr(,"class")
# [1] "histogram"

how to normalize a barplot to normal distribution in R?

1 Answers