I'm formatting figures for a journal that requires that tick marks be inside the axis. For my first plot using base graphics, I've found that I can use this:
attach(mtcars)
plot(wt, mpg, tck = 0.02)
My second plot, however, uses raw.means.plot2 from the plotrix package, and I can't figure out how to get the tick marks to move to the inside of the x axis. I've looked through the plotrix documentation, but I can't seem to find anything on this. For example, I'd like to move the tick marks to the inside of both the x and y axes on this plot, but tck=0.02 only moves the tick marks on the y axis.
mtcars$ID <- seq(1,32)
library('plotrix')
raw.means.plot2(mtcars, col.id="ID",col.offset="cyl", col.x="am", col.value="mpg", tck=0.02)
If passing inside the function doesn't work, the next thing to try is a call to par before plotting.
library('plotrix')
mtcars$ID <- seq(1,32)
par(tck = .02)
raw.means.plot2(mtcars, col.id="ID",col.offset="cyl", col.x="am", col.value="mpg")
And note that after plotting par('tck') is still 0.02, so you can use
op <- par(no.readonly = TRUE)
par(stuff)
plot(stuff)
par(op)
to return to previous state. Or graphics.off() to clear all plots and reset to defaults
Turn off the xaxis and then add one manually
raw.means.plot2(mtcars, col.id="ID",col.offset="cyl", col.x="am", col.value="mpg", xaxis = FALSE, tck = 0.02)
axis(1, tck = 0.02)
Related
I have a problem with the x-axis labelling.
I created a lm model(first.stage) and plotted the residuals using the code:
plot(first.stage$residuals,ylab="Deviation",xlab="Year",type="l",col="blue")
Now I'd like to change the x-axis labelling independent from the x values.
More precisely, I want to show the years 1960-2010 on the x axis. I tried a lot nothing worked.
If you want to set custom tick marks for your x axis, set xaxt = 'n' in the plot() call, then use axis(1, at = c()) to set the new tick marks. For example,
plot(mtcars$wt, mtcars$mpg, xaxt='n')
axis(1, at = c(1.5, 2.5))
I'm constructing a plot using bargraph.CI from sciplot. The x-axis represents a categorical variable, so the values of this variable are the names for the different positions on the x-axis. Unfortunately these names are long, so at default settings, some of them just disappear. I solved this problem by splitting them into multiple lines by injecting "\n" where needed. This basically worked, but because the names are now multi-line, they look too close to the x-axis. I need to move them farther away. How?
I know I can do this with mgp, but that affects the y-axis too.
I know I can set axisnames=FALSE in my call to barplot.CI, then use axis to create a separate x-axis. (In fact, I'm already doing that, but only to make the x-axis extend farther than it would by default- see my code below.) Then I could give the x-axis its own mgp parameter that would not affect the y-axis. But as far as I can tell, axis() is well set up for ordinal or continuous variables and doesn't seem to work great for categorical variables. After some fiddling, I couldn't get it to put the names in the right locations (i.e. right under their correspondence bars)
Finally, I tried using mgp.axis.labels from Hmisc to set ONLY the x-axis mgp, which is precisely what I want, but as far as I could tell it had no effect on anything.
Ideas? Here's my code.
ylim = c(0.5,0.8)
yticks = seq(ylim[1],ylim[2],0.1)
ylab = paste(100*yticks,"%",sep="")
bargraph.CI(
response = D$accuracy,
ylab = "% Accuracy on Test",
ylim = ylim,
x.factor = D$training,
xlab = "Training Condition",
axes = FALSE
)
axis(
side = 1,
pos = ylim[1],
at = c(0,7),
tick = TRUE,
labels = FALSE
)
axis(
side = 2,
tick = TRUE,
at = yticks,
labels = ylab,
las = 1
)
axis works fine with cateory but you should set the right ticks values and play with pos parameter for offset translation. Here I use xvals the return value of bargraph.CI to set àxis tick marks.
Here a reproducible example:
library(sciplot)
# I am using some sciplot data
dat <- ToothGrowth
### I create along labels
labels <- c('aaaaaaaaaa\naaaaaaaaaaa\nhhhhhhhhhhhhhhh',
'bbbbbbbbbb\nbbbbbbbbbbb\nhhhhhhhhhhhhhh',
'cccccccccc\nccccccccccc\ngdgdgdgdgd')
## I change factor labels
dat$dose <- factor(dat$dose,labels=labels)
ll <- bargraph.CI(x.factor = dose, response = len, data = dat,axisnames=FALSE)
## set at to xvals
axis(side=1,at=ll$xvals,labels=labels,pos=-2,tick=FALSE)
How do I plot tick marks inside of a lattice plot, opposed to the outside. "tcl = 0.5" seems to work for a normal plot; however not for my lattice dotplot.
Example of what I'd like to replicate in Lattice:
Thanks!
Tick length is controlled by the tck= argument, which needs to be passed in to higher-level plotting functions via their scales= argument. Setting tck=0 suppresses drawing of the ticks, while setting it to a negative number causes ticks to be drawn inside of the plot:
library(lattice)
xyplot(1:10 ~ 1:10,
scales = list(y = list(tck=0),
x = list(tck=c(-1, 0))))
How can I go about removing the box around an xyplot, while keeping the axis scale tick marks? In the spirit of Edward Tufte's minimalist data graphic aesthetic, these axis lines are "non-data ink," and can (should?) be "erased."
library(lattice)
my.df <- data.frame(x=-10:10)
my.df$y <- my.df$x^2
xyplot(y~x,data=my.df)
It seems that the trellis display parameters (e.g. axis.line$col) control both the axis lines and axis ticks together:
xyplot(y~x,data=my.df,
par.settings=list(axis.line=list(col="transparent")))
...which is not the desired result, so it doesn't look like there's a simple way to turn off the lines while leaving the box.
The best I've been able to come up with is a brute-force hack, where I build the tick marks by hand using panel.segments:
at.x=pretty(c(-10,10))
at.y=pretty(c(0,100))
xyplot(y~x,data=my.df,
par.settings=list(axis.line=list(col="transparent")),
scales=list(x=list(at=at.x,labels=at.x),
y=list(at=at.y,labels=at.y)),
panel=function(...){
panel.xyplot(...)
panel.segments(x0=at.x,x1=at.x,y0=-4,y1=-2)
panel.segments(x0=-11.5,x1=-11,y0=at.y,y1=at.y)
}
)
This is close to the desired result, but there's quite a bit of fiddling required to get the tick marks to be a reasonable length and offset a "nice" distance from the data points. These values won't translate from one graphic to the next. Plus, note that the axis labels are now padded too far from the tick marks. I'm sure there's a way to reduce that padding, but that would only make the code even uglier and less portable.
So how can one go about suppressing just the lines that make up the "box" around the plot area, while leaving the tick marks and axis labels intact? Bonus points if this approach could also be used to suppress some, but not all of the lines (e.g. leave the left and lower lines, but suppress the top and right lines).
This is still a bit hacky, but at least you don't have to do any figuring by hand. It uses a combination of par.settings and a custom axis function that takes an argument line.col and temporarily changes the axis line color by a call to trellis.par.set:
EDIT (removed unnecessary changing of trellis settings)
xyplot(y~x,data=my.df, par.settings = list(axis.line = list(col = "transparent")),
# Pass custom axis function to argument 'axis'
axis = function(side, line.col = "black", ...) {
# Only draw axes on the left and bottom
if(side %in% c("left","bottom")) {
# Call default axis drawing function
axis.default(side = side, line.col = "black", ...)
}
}
)
At the moment, I chalk up why line.col = "black" is required in the arguments of the custom axis function to magic. My guess is that it has to do with argument matching with the ellipses (...). Perhaps I'll be wiser tomorrow and find the true reason.
This results in:
The easiest thing to do is to use the custom axis function (axis). Just set lwd (line width) to zero and tick marks (lwd.ticks) to something else. It worked like a charm!
plot(NA,NA,type="n",xaxt="n", lwd=linewidth, xlim=c(1,24), xlab="", ylab="",ylim=c(-300,500))
axis(side = 4, tck = .05, **lwd=0, lwd.ticks=1**, line = 0, labels = NA, col= cols_border[1], col.axis = cols_black)
axis(side = 4, lwd = 0, line = -4.5, las = 1, cex.axis=axis_fontsize, col= cols_border[1], col.axis = cols_black)
mtext("Light deviations (lum/sec)",side=4, padj=-2.5, cex=title_fontsize, col="black")
I'm trying to generate a histogram in R with a logarithmic scale for y. Currently I do:
hist(mydata$V3, breaks=c(0,1,2,3,4,5,25))
This gives me a histogram, but the density between 0 to 1 is so great (about a million values difference) that you can barely make out any of the other bars.
Then I've tried doing:
mydata_hist <- hist(mydata$V3, breaks=c(0,1,2,3,4,5,25), plot=FALSE)
plot(rpd_hist$counts, log="xy", pch=20, col="blue")
It gives me sorta what I want, but the bottom shows me the values 1-6 rather than 0, 1, 2, 3, 4, 5, 25. It's also showing the data as points rather than bars. barplot works but then I don't get any bottom axis.
A histogram is a poor-man's density estimate. Note that in your call to hist() using default arguments, you get frequencies not probabilities -- add ,prob=TRUE to the call if you want probabilities.
As for the log axis problem, don't use 'x' if you do not want the x-axis transformed:
plot(mydata_hist$count, log="y", type='h', lwd=10, lend=2)
gets you bars on a log-y scale -- the look-and-feel is still a little different but can probably be tweaked.
Lastly, you can also do hist(log(x), ...) to get a histogram of the log of your data.
Another option would be to use the package ggplot2.
ggplot(mydata, aes(x = V3)) + geom_histogram() + scale_x_log10()
It's not entirely clear from your question whether you want a logged x-axis or a logged y-axis. A logged y-axis is not a good idea when using bars because they are anchored at zero, which becomes negative infinity when logged. You can work around this problem by using a frequency polygon or density plot.
Dirk's answer is a great one. If you want an appearance like what hist produces, you can also try this:
buckets <- c(0,1,2,3,4,5,25)
mydata_hist <- hist(mydata$V3, breaks=buckets, plot=FALSE)
bp <- barplot(mydata_hist$count, log="y", col="white", names.arg=buckets)
text(bp, mydata_hist$counts, labels=mydata_hist$counts, pos=1)
The last line is optional, it adds value labels just under the top of each bar. This can be useful for log scale graphs, but can also be omitted.
I also pass main, xlab, and ylab parameters to provide a plot title, x-axis label, and y-axis label.
Run the hist() function without making a graph, log-transform the counts, and then draw the figure.
hist.data = hist(my.data, plot=F)
hist.data$counts = log(hist.data$counts, 2)
plot(hist.data)
It should look just like the regular histogram, but the y-axis will be log2 Frequency.
I've put together a function that behaves identically to hist in the default case, but accepts the log argument. It uses several tricks from other posters, but adds a few of its own. hist(x) and myhist(x) look identical.
The original problem would be solved with:
myhist(mydata$V3, breaks=c(0,1,2,3,4,5,25), log="xy")
The function:
myhist <- function(x, ..., breaks="Sturges",
main = paste("Histogram of", xname),
xlab = xname,
ylab = "Frequency") {
xname = paste(deparse(substitute(x), 500), collapse="\n")
h = hist(x, breaks=breaks, plot=FALSE)
plot(h$breaks, c(NA,h$counts), type='S', main=main,
xlab=xlab, ylab=ylab, axes=FALSE, ...)
axis(1)
axis(2)
lines(h$breaks, c(h$counts,NA), type='s')
lines(h$breaks, c(NA,h$counts), type='h')
lines(h$breaks, c(h$counts,NA), type='h')
lines(h$breaks, rep(0,length(h$breaks)), type='S')
invisible(h)
}
Exercise for the reader: Unfortunately, not everything that works with hist works with myhist as it stands. That should be fixable with a bit more effort, though.
Here's a pretty ggplot2 solution:
library(ggplot2)
library(scales) # makes pretty labels on the x-axis
breaks=c(0,1,2,3,4,5,25)
ggplot(mydata,aes(x = V3)) +
geom_histogram(breaks = log10(breaks)) +
scale_x_log10(
breaks = breaks,
labels = scales::trans_format("log10", scales::math_format(10^.x))
)
Note that to set the breaks in geom_histogram, they had to be transformed to work with scale_x_log10