How to set limits of grid in barplot? - r

I am trying to put grids in my barplots, but they appear in front of the data and not in the background. I tried to fix this using
panel.first = grid()
As for the data I am trying to plot, the first column consists of the year numbers (2014-2021) and the second columns are the corresponding values (all vector classes are numeric). When trying to plot using the following code:
par(mfrow=c(1,2))
barplot(mean_trend[,2],names.arg = mean_trend[,1],col="skyblue",ylim = c(0.1,95),cex=0.8,cex.names=0.85,las=2,cex.lab=0.85,lwd=1.5,panel.first = grid())
mtext(side=2,line=2.3, "Average amount in mm", font=2, cex=0.8)
box(lwd=1.5)
barplot(freq_trend[,2],names.arg = freq_trend[,1],col="skyblue",ylim = c(2,4500),cex=0.85,cex.names=0.85,las=2,cex.lab=0.85,lwd=1.5,panel.first = grid())
box(lwd=1.5)
mtext(side=2,line=3.3, "Average flood frequency", font=2, cex=0.8)
I obtain the following result
As you can see, the grid is now behind the plotted data, but exceeds the box/plot limits. How can I fix this?
Kind regards

As you didn't add a data mean_trend - I give an example with other data.
About add and others arguments - you can read ?barplot
# One row, two columns
par(mfrow = c(1, 2))
#PLOT1
barplot(table(mtcars$cyl), main = "PLOT 1", col = c("yellow", "green", "red"), ylim = c(0, 15))
grid(nx = NULL, ny = NULL, lwd = 1, lty = 1, col = "gray")
barplot(table(mtcars$cyl), col = c("yellow", "green", "red"), ylim = c(0, 15), add = TRUE)
#PLOT2
barplot(table(mtcars$cyl), main = "PLOT 2", col = c("yellow", "green", "red"), ylim = c(0, 15))
grid(nx = NULL, ny = NULL, lwd = 1, lty = 1, col = "gray")
barplot(table(mtcars$cyl), col = c("yellow", "green", "red"), ylim = c(0, 15), add = TRUE)

Related

Save par(mfrow) multipanel output as svg

I'm using par(mfrow) to generate a multi-panel plot of three separate graph output objects. Sample code below represents a very simplified version of the objects I have.
How can I save these plots as a single object with ggsave? I've tried naming the par(mfrow) as an object and plotting it, but that doesn't seem to work.
Any advice on alternative ways of generating/saving a multi-panel plot is also welcome! Please let me know if I can clarify the question or example. Thanks!
par(mfrow = c(1,3), mar = c(10, 5, 5, 3), xpd = TRUE)
hist(x = rnorm(100), col = "skyblue", main = "X")
hist(x = rnorm(50), col = "green", main = "Y")
legend("bottom", c("Blue", "Green", "Purple"),
title = "Sample Data", horiz = TRUE, inset = c(0, -0.4),
col = c("skyblue", "green", "purple"), pch = rep(15,2),
bty = "n", pt.cex = 1.5, text.col = "black")
hist(x = rnorm(75), col = "purple", main = "Z")
I would suggest using svglite over svg if you want to edit the graph using Inkscape or a similar program since you will not be able to edit the text (change the text, font, or size) in a file produced by svg. Here is an example with a few edits to your original code:
library(svglite)
svglite("MyPlots.svg", width=8, height=6)
par(mfrow = c(1,3), mar = c(10, 5, 5, 3), xpd = TRUE, mgp=c(1.75, .75, 0))
hist(x = rnorm(100), col = "skyblue", main = "X")
hist(x = rnorm(50), col = "green", main = "Y")
legend("bottom", c("Blue", "Green", "Purple"),
title = "Sample Data", horiz = TRUE, inset = c(0, -0.2),
col = c("skyblue", "green", "purple"), pch = rep(15,2),
bty = "n", pt.cex = 1.5, text.col = "black")
hist(x = rnorm(75), col = "purple", main = "Z")
dev.off()

Changing colors of points using qqmath from the lattice package

I have a plot that I made using qqmath in the Lattice package (I subset it to only 3 points to make it easier for an example).
table <- data.table(Col1=c(12,3,4), Col2 = c(54,4,6), Col3 = c("Pink", "Pink", "Red"))
PrbGrd <- qnorm(c(0.00001,0.0001,0.001,0.01, 0.05, 0.10,0.20,0.30,0.40,
0.50, 0.60, 0.70,0.80,0.90,0.95,0.99,0.999,0.9999,0.99999))
PrbGrdL<-c("0.001","0.01","0.1","1","5","10","20","30","40","50","60","70","80","90","95","99","99.9","99.99","99.999")
PrbGrdL2<- c("99.999","99.99","99.9","99","95","90","80","70","60","50","40","30","20","10","5","1","0.1","0.01","0.001")
ValGrd<- c(seq(0.001,0.01,0.001),seq(0.01,0.1,0.01),seq(0.1,1,0.1),seq(1,10,1),seq(10,100,10),seq(100,1000,100),seq(1000,10000,1000))
ValGrd<- log10(ValGrd)
ValGrd2 <- c(-2:20)
ProbPlot <- qqmath(~ Col1,
data= table,
distribution = function(p) qnorm(p),
main = "Normal probability plot",
pch=20,
cex=0.5,
xlab="Probability of Lower",
ylab = "Pb",
#xlim = c(max(PrbGrd),min(PrbGrd)),
xlim = c(min(PrbGrd),max(PrbGrd)),
scales=list(y=list(alternating=1),x = list(at = PrbGrd, labels = PrbGrdL, cex = 0.8)),
#yscale.components=yscale.components.log10ticks,
panel=function(x,...){
panel.abline(v=PrbGrd ,col="grey",lty=3)
panel.abline(h=ValGrd2,col="grey",lty=3)
panel.qqmath(x,distribution=qnorm)
}
)
I would like to use the colors that are in the 3rd column of my table (Col3) to change the colors of the respective point on the plot. I can't figure out how to do it within qqmath, this would be simple with the regular plot function... but it doesn't seem as simple with qqmath.
Thank you!
Updated: based on OP's comment.
You can add the col argument, like:
ProbPlot <- qqmath(~ Col1,
data = table,
distribution = function(p) qnorm(p),
main = "Normal probability plot",
pch = 20,
cex = 0.5,
xlab = "Probability of Lower",
ylab = "Pb",
#xlim = c(max(PrbGrd),min(PrbGrd)),
xlim = c(min(PrbGrd),max(PrbGrd)),
scales = list(y = list(alternating = 1),
x = list(at = PrbGrd,
labels = PrbGrdL,
cex = 0.8)),
#yscale.components=yscale.components.log10ticks,
panel = function(x, ...){
panel.abline(v = PrbGrd,
col = "grey",
lty = 3)
panel.abline(h = ValGrd2,
col = "grey",
lty = 3)
panel.qqmath(x,
distribution=qnorm,
col = table$Col3) # add colors for each point
}
)

How to customise dates (years) on the x-axis in R

I am struggling to customise the jump size on the x-axis in R.
Current code:
par(mfrow = c(2,2))
r.star.ts.sp <- ts(r.star.sp, frequency = 4, start = c(1978,1), end = c(2018, 1))
# Big drop in r* around 123th quarter equivalent to 2008:Q4 / 2009:Q1
trendgrowth.ts.sp <- ts(trendgrowth.sp, frequency = 4, start = c(1978,1), end = c(2018, 1))
plot.ts(r.star.ts.sp,
ylim = c(-3, 4), xlab = " ", ylab = " ", axes = F, col = "blue")
lines(trendgrowth.ts.sp, lty = 2, col = "red")
abline(h = 0, lty = 2)
title(main ="r* and Trend Growth", line = 0.5, font.main = 3)
box()
axis(4)
axis(1)
legend("bottomleft", legend = c("r*", "Trend Growth (g)"),
bty = "n", lty = c(1,2), col = c("blue", "red"), horiz = F, text.col = "black",
cex = 1, pt.cex = .5, inset = c(0.02, 0.02))
# -------------------------------------- #
# Plot output gap and real rate gap
# -------------------------------------- #
outputgap.ts.sp <- ts(outputgap.sp, frequency = 4, start = c(1978,1), end = c(2018, 1))
realrategap.ts.sp <- ts(realrategap.sp, frequency = 4, start = c(1978,1), end = c(2018, 1))
plot.ts(outputgap.ts.sp, ylim = c(-20, 15), xlab=" ", ylab=" ", axes = F, col="blue")
lines(realrategap.ts.sp, lty = 2, col = "red")
abline(h = 0, lty = 2)
legend("topright", legend = c("Output Gap", "Real Rate Gap"),
bty = "n", lty = c(1,2), col = c("blue", "red"), horiz = F, text.col = "black",
cex = 1, pt.cex = .5, inset = c(0.02, 0.02))
title(main = "Output Gap and Real Rate Gap", line = 0.5, font.main = 3)
box()
axis(side = 4)
axis(side = 1)
How would one specify the years on the x-axis from 1975 to 2020 with jumps of 5 years?
Furthermore, (off-topic) I need two plots next to each other, but I feel that par(mfrow = c(2,2)) is not the correct statement. However, changing it into par(mfrow = c(1,2)) creates abnormal large figures.
Thanks!
The OP has requested to specify the years on the x-axis from 1975 to 2020 with jumps of 5 years.
This can be achieved by
axis(1, at = seq(1975L, 2020L, by = 5L))
However, the result may depend on the mfrow parameter. Here is a a dummy example using par(mfrow = c(2, 2)):
Note that the x-axis of the left graph was created by axis(1) while the x-axis of the right graph was created by axis(1, at = seq(1975L, 2020L, by = 5L)). Also note the large white space below the two graphs.
With par(mfrow = c(1, 2)) the result becomes
Here, the right graph shows unlabeled ("minor") tick marks. This is explained in the mfrow section of ?par: In a layout with exactly two rows and columns the base value of "cex" is reduced by a factor of 0.83. So, font size is reduzed by 17% per cent which allows to label all tick marks without overplotting.

How to remove the zero labels in Histogram plot in R?

Any tips to remove the zero labels in between the histogram bars?
hist(links$Survey_Duration, breaks = seq(0,50,5), main = "Survey Duration",
labels = TRUE, border = "black",
xlab = "Survey", ylim = c(0, 15), col = "gray", las = 1, xaxt='n')
axis(side=1, at=seq(0,50,5), labels=seq(0,50,5))
abline(v = mean(links$Survey_Duration), col = "royalblue", lwd = 1.5)
abline(v = median(links$Survey_Duration), col = "red", lwd = 1.5)
legend(x = "topright", c("Mean", "Median"), col = c("royalblue","red"),
lwd = c(1.5,1.5))
How about this?
# modify data so there's zero in one of the bins
mtcars$mpg <- ifelse(mtcars$mpg >= 25 & mtcars$mpg <= 30, NA, mtcars$mpg)
# save plot parameters
h <- hist(mtcars$mpg, plot = FALSE)
# produce plot
plot(h, ylim = c(0, 14))
# add labels manually, recoding zeros to nothing
text(h$mids, h$counts + 1, ifelse(h$counts == 0, "", h$counts))
A slightly different answer using the labeling in hist instead of adding text afterwards.
You do not provide your data, so I will use some data that is handy to illustrate.
The labels argument can specify the individual labels
H1 = hist(iris$Sepal.Length, breaks = 3:8, plot=FALSE)
BarLabels = H1$counts
BarLabels[BarLabels == 0] = ""
hist(iris$Sepal.Length, breaks = 3:8, labels = BarLabels)
Thanks #Daniel Anderson, it Ok now (Thumbs Up)
links$Survey_Duration <- ifelse(links$Survey_Duration > 15 &
links$Survey_Duration <= 25,
NA,
links$Survey_Duration)
h <- hist(links$Survey_Duration, breaks = seq(0,50,5), plot = FALSE)
plot(h, ylim = c(0, 14), main = "Survey Duration", xlab = "Time", col = "gray", las = 1)
text(h$mids, h$counts + 1, ifelse(h$counts == 0, "", h$counts))
axis(side=1, at=seq(0,50,5), labels=seq(0,50,5))
abline(v = mean(links$Survey_Duration), col = "royalblue", lwd = 1.5)
abline(v = median(links$Survey_Duration), col = "red", lwd = 1.5)
legend(x = "topright",
c("Mean", "Median"),
col = c("royalblue","red"),
lwd = c(1.5,1.5))

R - Plot with two y axis, bars and points not aligned

For a customer I'm trying to do a combined barplot and lineplot (with points) with two y axis.
Problem: My bars and points are not aligned.
Background: We have several machines and are measuring their number of on/of switches and the amount of time that each machine is running. We want both information together in one plot to save space, because we have several machines.
The data is aggregated by day or hour. Here's some sample data:
date <- seq(as.Date("2016-10-01"), as.Date("2016-10-10"), "day")
counts <- c(390, 377, 444, NA, NA, NA, NA, 162, 166, 145)
runtime <- c(56.8, 59.4, 51.0, NA, NA, NA, NA, 38.5, 40.9, 43.4)
df <- data.frame(date = date, counts = counts, runtime = runtime)
Here's what I tried so far:
par(mar = c(3,4,4,4) + 0.3)
barplot(df$runtime, col = "palegreen2", border = "NA", ylab = "runtime in [%]",
ylim = c(0,100), font.lab = 2)
par(new = TRUE)
ymax <- max(df$counts, na.rm = TRUE) * 1.05
plot(df$date, df$counts, type = "n", xlab = "", ylab = "", yaxt = "n",
main = "Machine 1", ylim = c(0, ymax))
abline(v = date, col = "red", lwd = 2.5)
lines(df$date, df$counts, col = "blue", lwd = 2)
points(df$date, df$counts, pch = 19, cex = 1.5)
axis(4)
mtext("Number of switching operations", side = 4, line = 3, font = 2)
I found some inspiration for two axis here: http://robjhyndman.com/hyndsight/r-graph-with-two-y-axes/
What can I do to get bars with their middle aligned with the points of the lineplot?
The problem you are running into is the call to the second plot function after the barplot. This is shifting/resizing the plotting canvas which is causing the shift in the subsequent points.
Here is a quick work-around that just rescales the points and lines onto the barplot. It saves the barplot as an object, which stores x-axis locations for the mid-points of the bars. Then, when you plot the abline, lines and points using 'bp' as the x-axis variable, they will be correctly aligned.
ymax <- max(df$counts, na.rm = TRUE) * 1.05
par(mar=c(4.1,5.1,2.1,5.1))
bp <- barplot(df$runtime, col = "palegreen2", border = "NA", ylab = "runtime in [%]",
ylim = c(0,100), font.lab = 2, xlim=c(0.2,12), )
barplot(df$runtime, col = "palegreen2", ylab = "runtime in [%]", border="NA",
ylim = c(0,100), font.lab = 2)
abline(v = bp, col = "red", lwd = 2.5)
lines(bp, df$counts/ymax*100, col = "blue", lwd = 2)
points(bp, df$counts/ymax*100, pch = 19, cex = 1.5)
axis(4,at=c(0,20,40,60,80,100), labels=c("0","100","200","300","400","500"))
mtext("Number of switching operations", side = 4, line = 3, font = 2)
axis(1, at=bp, labels=df$date)
#emilliman: Thank you for your patience and input! Your plot is not completely correct, because the scaling of the second y-axis does not fit the points' values, but your idea helped me to find a solution!
Here's my new code:
library(plyr)
ymax <- max(df$counts, na.rm = TRUE)
ymax_up <- round_any(ymax, 100, f = ceiling)
ylab <- ymax_up/5 * c(0:5)
par(mar = c(3,4,4,4) + 0.3)
bp <- barplot(df$runtime, col = "palegreen2", border = "NA", ylab = "runtime in [%]",
ylim = c(0,100), font.lab = 2, main = "Machine 1")
abline(v = bp, col = "red", lwd = 2.5)
lines(bp, 100/ymax_up * df$counts, col = "blue", lwd = 2)
points(bp, 100/ymax_up * df$counts, pch = 19, cex = 1.5)
axis(4,at=c(0,20,40,60,80,100), labels= as.character(ylab))
mtext("Number of switching operations", side = 4, line = 3, font = 2)
xlab <- as.character(df$date, format = "%b %d")
axis(1, at=bp, labels = xlab)
abline(h = c(0,100))
(http://i.imgur.com/9YtYGSD.png)
Maybe this is helpful for others who run into this problem.

Resources