I know how to plot several density curves/polygrams on one plot, but not conditional density plots.
Reproducible example:
require(ggplot2)
# generate data
a <- runif(200, min=0, max = 1000)
b <- runif(200, min=0, max = 1000)
c <- sample(c("A", "B"), 200, replace =T)
df <- data.frame(a,b,c)
# plot 1
ggplot(df, aes(a, fill = c)) +
geom_density(position='fill', alpha = 0.5)
# plot 2
ggplot(df, aes(b, fill = c)) +
geom_density(position='fill', alpha = 0.5)
In my real data I have a bunch of these paired conditional density plots and I would need to overlay one over the other to see (and show) how different (or similar) they are. Does anyone know how to do this?
One way would be to plot the two versions as layers. The overlapping areas will be slightly different, depending on the layer order, based on how alpha works in ggplot2. This may or may not be what you want. You might fiddle with the two alphas, or vary the border colors, to distinguish them more.
ggplot(df, aes(fill = c)) +
geom_density(aes(a), position='fill', alpha = 0.5) +
geom_density(aes(b), position='fill', alpha = 0.5)
For example, you might make it so the fill only applies to one layer, but the other layer distinguishes groups using the group aesthetic, and perhaps a different linetype. This one seems more readable to me, especially if there is a natural ordering to the two variables that justifies putting one in the "foreground" and one in the "background."
ggplot(df) +
geom_density(aes(a, group = c), position='fill', alpha = 0.2, linetype = "dashed") +
geom_density(aes(b, fill = c), position='fill', alpha = 0.5)
I'm not so sure if "on top of one another" is a great idea. Jon's ideas are probably the way to go. But what about just plotting side-by side - our brains can cope with that and we can compare this pretty well.
Make it long, then use facet.
Another option might be an animated graph (see 2nd code chunk below).
require(ggplot2)
#> Loading required package: ggplot2
library(tidyverse)
a <- runif(200, min=0, max = 1000)
b <- runif(200, min=0, max = 1000)
#### BAAAAAD idea to call anything "c" in R!!! Don't do this. ever!
d <- sample(c("A", "B"), 200, replace =T)
df <- data.frame(a,b,d)
df %>% pivot_longer(cols = c(a,b)) %>%
ggplot(aes(value, fill = d)) +
geom_density(position='fill', alpha = 0.5) +
facet_grid(~name)
library(gganimate)
p <- df %>% pivot_longer(cols = c(a,b)) %>%
ggplot(aes(value, fill = d)) +
geom_density(position='fill', alpha = 0.5) +
labs(title = "{closest_state}")
p_anim <- p + transition_states(name)
animate(p_anim, duration = 2, fps = 5)
Created on 2022-06-14 by the reprex package (v2.0.1)
Although it is not the overlay you might have thought of, it facilitates the comparison of density curves:
library(tidyverse)
library(ggridges)
library(truncnorm)
DF <- tibble(
alpha = rtruncnorm(n = 200, a = 0, b = 1000, mean = 500, sd = 50),
beta = rtruncnorm(n = 200, a = 0, b = 1000, mean = 550, sd = 50)
)
DF <- DF %>%
pivot_longer(c(alpha, beta), names_to = "name", values_to = "meas") %>%
mutate(name = factor(name))
DF %>%
ggplot(aes(meas, name, fill = factor(stat(quantile)))) +
stat_density_ridges(
geom = "density_ridges_gradient",
calc_ecdf = T,
quantiles = 4,
quantile_lines = T
) +
scale_fill_viridis_d(name = "Quartiles")
Related
I'm currently finishing off my Masters project and need to include some graphics for the write-up. Without boring you too much, I have some data which is associated with AR(1) parameters ranging from 0.1 to 0.9 by 0.1 increments. As such I thought of doing a faceted histogram like the one below (worry not about the hideous fruit salad of colours, it will not be used).
I used this code.
ggplot(opt_lens_geom,aes(x=l_1024,fill=factor(rho))) + geom_histogram()+coord_flip()+facet_grid(.~rho,scales = "free_x")
I also would like to draw a trend line for the median values since the AR(1) parameter is continuous. In a later iteration I deleted the padding and made it "look" like it was one graph, but I have had issues with the endpoints matching up since each facet is a separate graphical device. Can anyone give me some advice on how to do this? I am not particularly partial to the faceting so if it is not needed I do away with it.
I will try and upload sample data, but all simulating 100 values for each of the 9 rhos would work just to get it started like:
opt_lens_geom <- data.frame(rho= rep(seq(0.1,0.9,by=0.1),each=100),l_1024=rnorm(900))
You might consider ggridges. I've assumed here that you want a median value for each value of rho.
library(ggplot2)
library(ggridges)
library(dplyr)
set.seed(1001)
opt_lens_geom <- data.frame(rho = rep(seq(0.1, 0.9, by = 0.1), each = 100),
l_1024 = rnorm(900))
opt_lens_geom %>%
mutate(rho_f = factor(rho)) %>%
ggplot(aes(l_1024, rho_f)) +
stat_density_ridges(quantiles = 2, quantile_lines = TRUE)
Result. You can add scale = 1 as a parameter to stat_density_ridges if you don't like the amount of overlap.
Try the following. It uses a pre-computed data frame of the medians.
library(ggplot2)
df <- iris[c(1, 5)]
names(df) <- c("val", "rho")
med <- plyr::ddply(df, "rho", summarise, m = median(val))
ggplot(data = df, aes(x = val, fill = factor(rho))) +
geom_histogram() +
coord_flip() +
geom_vline(data = med, aes(xintercept = m), colour = 'black') +
facet_wrap(~ factor(rho))
You could do a variant on this using geom_violin instead of using histograms, although you wouldn't get labelled counts, just an idea of the relative density. Example with made up data:
df = data.frame(
rho = rep(c(0.1, 0.2, 0.3), each = 50),
val = sample(1:10, 150, replace = TRUE)
)
df$val = df$val + (5 * (df$rho == 0.2)) + (8 * (df$rho == 0.3))
ggplot(df, aes(x = rho, y = val, fill = factor(rho))) +
geom_violin() +
stat_summary(aes(group = 1), colour = "black",
geom = "line", fun.y = "median")
This produces a violin for each value of rho, and joins the medians for each violin.
I want to create a plot using facet_grid(), with free scales for the y axis. However, for each row, the scale breaks should be distributed evenly, that is, with 3 breaks.
I lended from this question, but I was not able to adapt the code in a way that the scale breaks are actually pretty.
However, this is my current approach:
# Packages
library(dplyr)
library(ggplot2)
library(scales)
# Test Data
set.seed(123)
result_df <- data.frame(
variable = rep(c(1,2,3,4), each = 4),
mode = rep(c(1,2), each = 2),
treat = rep(c(1,2)) %>% as.factor(),
mean = rnorm(16, mean = .7, sd = 0.2),
x = abs(rnorm(16, mean = 0, sd = 0.5))) %>%
mutate(lower = mean - x,upper = mean + x)
# Function for equal breaks, lended from
equal_breaks <- function(n = 3, s = 0.05, ...) {
function(x) {
d <- s * diff(range(x)) / (1+2*s)
round(seq(min(x)+d, max(x)-d, length=n), 2)
}}
## Plot
result_df %>%
ggplot(aes(y = mean*100, x = treat)) +
geom_pointrange(aes(ymin = lower*100, ymax = upper*100), shape = 20) +
facet_grid(variable ~ mode, scales = "free_y")+
scale_y_continuous(breaks = equal_breaks(n = 3, s = .2))+
labs(x = "", y = "")
Which leads to this current plot. As one can see, the breaks are far from being reasonable.
Thanks in advance for any kind of recommendation, and please excuse me in case I have missed a already existing solution.
Best, Malte
I wonder if there is the possibility to change the fill main colour according to a categorical variable
Here is a reproducible example
df = data.frame(x = c(rnorm(10, mean = 0),
rnorm(10, mean = 3)),
y = c(rnorm(10, mean = 0),
rnorm(10, mean = 3)),
grp = c(rep('a', times = 10),
rep('b', times = 10)),
val = rep(1:10, times = 2))
ggplot(data = df,
aes(x = x,
y = y)) +
geom_point(pch = 21,
aes(color = grp,
fill = val,
size = val))
Of course it is easy to change the circle colour/shape, according to the variable grp, but I'd like to have the a group in shades of red and the b group in shades of blue.
I also thought about using facets, but don't know if the fill gradient can be changed for the two panels.
Anyone knows if that can be done, without gridExtra?
Thanks!
I think there are two ways to do this. The first is using the alpha aesthetic for your val column. This is a quick and easy way to accomplish your goal but may not be exactly what you want:
ggplot(data = df,
aes(x = x,
y = y)) +
geom_point(pch = 21,
aes(alpha=val,
fill = grp,
size = val)) + theme_minimal()
The second way would be to do something similar to this post: Vary the color gradient on a scatter plot created with ggplot2. I edited the code slightly so its not a range from white to your color of interest but from a lighter color to a darker color. This requires a little bit of work and using the scale_fill_identity function which basically takes a variable that has the colors you want and maps them directly to each point (so it doesn't do any scaling).
This code is:
#Rescale val to [0,1]
df$scaled_val <- rescale(df$val)
low_cols <- c("firebrick1","deepskyblue")
high_cols <- c("darkred","deepskyblue4")
df$col <- ddply(df, .(grp), function(x)
data.frame(col=apply(colorRamp(c(low_cols[as.numeric(x$grp)[1]], high_cols[as.numeric(x$grp)[1]]))(x$scaled_val),
1,function(x)rgb(x[1],x[2],x[3], max=255)))
)$col
df
ggplot(data = df,
aes(x = x,
y = y)) +
geom_point(pch = 21,
aes(
fill = col,
size = val)) + theme_minimal() +scale_fill_identity()
Thanks to this other post I found a way to visualize the fill bar in the legend, even though that wasn't what I meant to do.
Here's the ouptup
And the code
df = data.frame(x = c(rnorm(10, mean = 0),
rnorm(10, mean = 3)),
y = c(rnorm(10, mean = 0),
rnorm(10, mean = 3)),
grp = factor(c(rep('a', times = 10),
rep('b', times = 10)),
levels = c('a', 'b')),
val = rep(1:10, times = 2)) %>%
group_by(grp) %>%
mutate(scaledVal = rescale(val)) %>%
ungroup %>%
mutate(scaledValOffSet = scaledVal + 100*(as.integer(grp) - 1))
scalerange <- range(df$scaledVal)
gradientends <- scalerange + rep(c(0,100,200), each=2)
ggplot(data = df,
aes(x = x,
y = y)) +
geom_point(pch = 21,
aes(fill = scaledValOffSet,
size = val)) +
scale_fill_gradientn(colours = c('white',
'darkred',
'white',
'deepskyblue4'),
values = rescale(gradientends))
Basically one should rescale fill values (e.g. between 0 and 1) and separate them using another order of magnitude, provided by the categorical variable grp.
This is not what I wanted though: the snippet can be improved, of course, to make the whole thing less manual, but still lacks the simple usual discrete fill legend.
I came across what I found to be some helpful figures by David L. Stern at howtogiveatalk.com that have a jitterplot next to a mean/sd summary for each categorical variable.
Here is the first example from the page linked above
I spent some time trying to find similar figures online and couldn't.
I'm not sure which software and packages he used to create these figures (UPDATE: David Stern responded and explained that he uses a custom built Matlab function). I am most familiar with R and ggplot2 and figure it must be possible to create something similar using these tools. I tried to jump right in and make it but can't figure out where to go from here.
How I got started:
library(dplyr)
library(ggplot2)
library(tidyr)
df <- data_frame(a = rnorm(100, mean = 0.75, sd = 0.5), b = rgamma(100, shape = 0.75, scale = 0.5), c = rbinom(100, size = 1, prob = 0.6))
df <- gather(df)
df.sum <- df %>% group_by(key) %>% summarise(mean = mean(value), sd = sd(value))
ggplot(data = df.sum, aes(x = key)) +
geom_jitter(data = df, aes(y = value)) +
geom_point(aes(y = mean)) +
geom_linerange(aes(x = key, y = mean, ymin = (mean - sd), ymax = (mean + sd))) +
theme_bw()
Which produces the following graph:
The code is pretty rough, but gets most of the way there. I can't figure out how to move the geom_point and geom_linerange beside the jitter, though.
So how can this figure be made in R (preferably using ggplot2)?
I've figured it out! I'll post the answer here for future reference and for anyone else wanting to make a similar plot.
The key for me came down to converting the x-axis from a factor to a numeric in order to apply the shift.
library(dplyr)
library(ggplot2)
library(tidyr)
set.seed(125)
df <- data_frame(Normal = rnorm(100, mean = 0.5, sd = 0.5),
Gamma = rgamma(100, shape = 0.5, scale = 0.5),
Bimodal = c(rnorm(50, mean = 0.1, sd = 0.15), rnorm(50, mean = 0.9, sd = 0.15))
)
df <- gather(df)
df.sum <- df %>%
group_by(key) %>%
summarise(mean = mean(value), sd = sd(value))
ggplot(data = df, aes(x = key, y = value)) +
geom_jitter(position = position_jitter(width = 0.2), shape = 1, size = 3.5) +
geom_pointrange(data = df.sum, aes(x = as.numeric(key)+0.3, y = mean, ymin = (mean - sd), ymax = (mean + sd))) +
geom_point(data = df.sum, aes(x = as.numeric(key)+0.3, y = mean), size = 3.5) +
theme_bw() + xlab("") + ylab("Arbitrary Units")
It would be great if this code could be adapted into a ggplot extension to make this into a simple geom. I might take on the challenge myself if I can find the time.
Fairly straightforward without ggplot2
x<-0.5+runif(100,-0.2,0.2)
y<-rbind(rnorm(100,1,1),rgamma(100,1,1),rbinom(100,1,0.5)*2+rnorm(100,0,0.2))
for (j in 0:2){
if (j==0){plot(x,y[1,],xlim=c(0,4),ylim=c(-1,5),xlab="",ylab="Arbitrary Units",xaxt="n",bty="n",col="gray50")}
else{points(x+j, y[j+1,],col="gray50")}
points(j+0.9, mean(y[j+1,]),pch=19)
arrows(j+0.9,mean(y[j+1,])-sd(y[j+1,]),j+0.9,mean(y[j+1,])+sd(y[j+1,]), angle=90,length=0)
} # for j categories
axis(1,seq(0.5,2.5, by=1),tick=F,labels=c("Normal","Gamma","Bimodal"))
I am making boxplots with ggplot with data that is classified by 2 factor variables. I'd like to have the box sizes reflect sample size via varwidth = TRUE but when I do this the boxes overlap.
1) Some sample data with a 3 x 2 structure
data <- data.frame(group1= sample(c("A","B","C"),100, replace = TRUE),group2= sample(c("D","E"),100, replace = TRUE) ,response = rnorm(100, mean = 0, sd = 1))
2) Default boxplots: ggplot without variable width
ggplot(data = data, aes(y = response, x = group1, color = group2)) + geom_boxplot()
I like how the first level of grouping is shown.
Now I try to add variable widths...
3) ...and What I get when varwidth = TRUE
ggplot(data = data, aes(y = response, x = group1, color = group2)) + geom_boxplot(varwidth = T)
This overlap seems to occur whether I use color = group2 or group = group2 in both the main call to ggplot and in the geom_boxplot statement. Fussing with position_dodge doesn't seem to help either.
4) A solution I don't like visually is to make unique factors by combining my group1 and group2
data$grp.comb <- paste(data$group1, data$group2)
ggplot(data = data, aes(y = response, x = grp.comb, color = group2)) + geom_boxplot()
I prefer having things grouped to reflect the cross classification
5) The way forward:
I'd like to either a)figure out how to either make varwidth = TRUE not cause the boxes to overlap or b)manually adjusted the space between the combined groups so that boxes within the 1st level of grouping are closer together.
I think your problem can be solved best by using facet_wrap.
library(ggplot2)
data <- data.frame(group1= sample(c("A","B","C"),100, replace = TRUE), group2=
sample(c("D","E"),100, replace = TRUE) ,response = rnorm(100, mean = 0, sd = 1))
ggplot(data = data, aes(y = response, x = group2, color = group2)) +
geom_boxplot(varwidth = TRUE) +
facet_wrap(~group1)
Which gives:
A recent update to ggplot2 makes it so that the code provided by #N Brouwer in (3) works as expected:
# library(devtools)
# install_github("tidyverse/ggplot2")
packageVersion("ggplot2") # works with v2.2.1.9000
library(ggplot2)
set.seed(1234)
data <- data.frame(group1= sample(c("A","B","C"), 100, replace = TRUE),
group2= sample(c("D","E"), 100, replace = TRUE),
response = rnorm(100, mean = 0, sd = 1))
ggplot(data = data, aes(y = response, x = group1, color = group2)) +
geom_boxplot(varwidth = T)
(I'm a new user and can't post images inline)
fig 1
This question has been answered here ggplot increase distance between boxplots
The answer involves using the position = position_dodge() argument of geom_boxplot().
For your example:
data <- data.frame(group1= sample(c("A","B","C"),100, replace = TRUE), group2=
sample(c("D","E"),100, replace = TRUE) ,response = rnorm(100, mean = 0, sd = 1))
ggplot(data = data, aes(y = response, x = group1, color = group2)) +
geom_boxplot(position = position_dodge(1))