Issue with Density Plot using GGPLOT2 [closed] - r

Question

Closed. This question needs details or clarity. It is not currently accepting answers.
Want to improve this question? Add details and clarify the problem by editing this post.
Closed 3 years ago.
Improve this question
I want to plot a density plot for 2 groups and below is my code.
library(ggplot2)
#Sample data
dat <- data.frame(Score = c(myfiles2Best$V2, myfilesL2Best$V2)
, Group = rep(c("T", "L")))
ggplot(dat, aes(x = Score)) +
geom_density(aes(color = Group)) + xlim(0,16)
Below is the image of the output.
and when I change the data frame by changing the location of the column as shown below this is how my plot looks like.
dat <- data.frame(Score = c(myfilesL2Best$V2, myfiles2Best$V2)
, Group = rep(c("L", "T")))
Individually, this is how they look like.
dat <- data.frame(Score = c(myfiles2Best$V2)
, Group = rep(c("T"))
ggplot(dat, aes(x = Score)) +
geom_density(aes(color = Group)) + xlim(0,16)
dat <- data.frame(Score = c(myfilesL2Best$V2)
, Group = rep(c("L"))
ggplot(dat, aes(x = Score)) +
geom_density(aes(color = Group)) + xlim(0,16)
This is totally wrong, anything wrong with my setup
rownumber score group
1 8 T
2 8 L
3 7 T
4 7 L
5 9 T
6 8 L
7 8 T
8 7 L
9 8 T
10 8 L
11 8 T
12 9 L
13 8 T
14 8 L
15 8 T
16 8 L
17 9 T
18 7 L
19 9 T
20 7 L
21 8 T
22 10 L
23 8 T
24 8 L
25 9 T
26 8 L
27 8 T
28 8 L
29 9 T
30 8 L
31 7 T
32 10 L
33 8 T
34 10 L
35 8 T
36 7 L
37 8 T
38 7 L
39 11 T
40 9 L
41 8 T
42 9 L
43 8 T
44 10 L
45 8 T
46 9 L
47 8 T
48 8 L
49 8 T
50 7 L
51 9 T
52 8 L
53 8 T
54 9 L
55 8 T
56 7 L
57 7 T
58 9 L
59 10 T
60 8 L

Answer 1

ggplot2::geom_density uses the base R density function to compute density. (see ?geom_density.) This requires a parameter for smoothing, which by default uses a rule named "nrd0", which was picked for "historical and compatibility reasons." (see ?density.) You will get density plots with different appearances depending on this parameter.
From ?bandwidth:
bw.nrd0 implements a rule-of-thumb for choosing the bandwidth of a Gaussian kernel density estimator. It defaults to 0.9 times the minimum of the standard deviation and the interquartile range divided by 1.34 times the sample size to the negative one-fifth power (= Silverman's ‘rule of thumb’, Silverman (1986, page 48, eqn (3.31))) unless the quartiles coincide when a positive result will be guaranteed.
In your example, the two subgroups look like they have different standard deviations and IQRs, so it makes sense to me that they would look different depending on whether that smoothing parameter is calculated for them collectively (as in the case with the combined plot) or individually.
If you want your density plots to correspond between a grouped and individual basis, specify the bandwidth manually:
ggplot(df, aes(x = score)) +
geom_density(aes(color = group), bw = 0.3) +
xlim(0,16)
ggplot(subset(df, group == "L"), aes(x = score)) +
geom_density(aes(color = group), bw = 0.3) +
xlim(0,16)
ggplot(subset(df, group == "T"), aes(x = score)) +
geom_density(aes(color = group), bw = 0.3) +
xlim(0,16)