I have a sample dataset
d=data.frame(n=rep(c(1,1,1,1,1,1,2,2,2,3),2),group=rep(c("A","B"),each=20),stringsAsFactors = F)
And I want to draw two separate histograms based on group variable.
I tried this method suggested by #jenesaisquoi in a separate post here
Generating Multiple Plots in ggplot by Factor
ggplot(data=d)+geom_histogram(aes(x=n,y=..count../sum(..count..)),binwidth = 1)+facet_wrap(~group)
It did the trick but if you look closely, the proportions are wrong. It didn't calculate the proportion for each group but rather a grand proportion. I want the proportion to be 0.6 for number 1 for each group, not 0.3.
Then I tried dplyr package, and it didn't even create two graphs. It ignored the group_by command. Except the proportion is right this time.
d%>%group_by(group)%>%ggplot(data=.)+geom_histogram(aes(x=n,y=..count../sum(..count..)),binwidth = 1)
Finally I tried factoring with color
ggplot(data=d)+geom_histogram(aes(x=n,y=..count../sum(..count..),color=group),binwidth = 1)
But the result is far from ideal. I was going to accept one output but with the bins side by side, not on top of each other.
In conclusion, I want to draw two separate histograms with correct proportions calculated within each group. If there is no easy way to do this, I can live with one graph but having the bins side by side, and with correct proportions for each group. In this example, number 1 should have 0.6 as its proportion.
By changing ..count../sum(..count..) to ..density.., it gives you the desired proportion
ggplot(data=d)+geom_histogram(aes(x=n,y=..density..),binwidth = 1)+facet_wrap(~group)
You actually have the separation of charts by variable correct! Especially with ggplot, you sometimes need to consider the scales of the graph separately from the shape. Facet_wrap applies a new layer to your data, regardless of scale. It will behave the same, no matter what your axes are. You could also try adding scale_y_log10() as a layer, and you'll notice that the overall shape and style of your graph is the same, you've just changed the axes.
What you actually need is a fix to your scales. Understandable - frequency plots can be confusing. ..count../sum(..count..)) treats each bin as an independent unit, regardless of its value. See a good explanation of this here: Show % instead of counts in charts of categorical variables
What you want is ..density.., which is basically the count divided by the total count. The difference is subtle in principle, but the important bit is that the value on the x-axis matters. For an extreme case of this, see here: Normalizing y-axis in histograms in R ggplot to proportion, where tiny x-axis values produced huge densities.
Your original code will still work, just substituting the aesthetics I described above.
ggplot(data=d)+geom_histogram(aes(x=n,y=..density..,)binwidth = 1)+facet_wrap(~group)
If you're still confused about density, so are lots of people. Hadley Wickham wrote a long piece about it, you can find that here: http://vita.had.co.nz/papers/density-estimation.pdf
Related
I've come up with a graph (a scatterplot) of the log(1+inf) (inf = number of people infected with a given disease on the y-axis against one of the explanatory variables, in this case, the populational density (pop./km²; x-axis) on my model. The log transformation was used merely for visualization, because it spreads the distribution of the data and allows for more aesthetically appealing plots. Basically, what I want is both axis to show the value of that same variable before the log transformation. The dots need to be plotted like plot(log(1+inf),log(populational_density), but the number on the axis should refer to plot(inf,populational_density). I've provided a picture of my graph with some manual editing on the y-axis to show you the idea of what I want.
The numbers in red would be the 'inf' values equivalent to log(inf);
Please, bear in mind that those values in red do not correspond to reality.
I understand the whole concept of y = f(x), but i've been asked to provide it. Is this possible? I'm using the ggplot2package for plotting.
I just came into a problem while making several maps in R, the problem I came to is that I want to plot several maps and some geom_points in those maps, each map will have some points with different values and so the legend with the scales (size and color) will change between maps. All I want is to have exactly the same legend, representing the same values (for both color and size). I've tried with breaks etc but my data is continuous, so I didn't find any way to fix it.
EDIT:Simple example
Will try to explain with simple example by myself. Imagine I have these two arrays to be plotted into different coordinates for 2 different days:
c<-(1,2,3,2,1)
c<-(1,9,2,1,2)
What I want is to set the legend of the plot to be always representing the range 1-9 as values of the geom_points, no matter the specific values of the given day, in a way that no matter the values, the legend will be always the same and if I try to set some slides, the scale will not change
Any ideas?
I want to plot some data with barplot. Rather, I want to make a bar graph and barplot seemed the logical choice. I am plotting just fine but I was wondering if there is a way to intelligently scale the y axis to round up from the highest count.
For example I set the yaxis in this case to be 30, because I knew that Strand.22 had 27 counts in it: barplot(unlist(d), ylim=c(0,30), xlab="Forward Reverse", ylab="Counts")
In the future, I want this script to run on its own, so it would be optimal for the the Y-axis to choose it's own ylim. Short of pulling the information out of my 'd' variable I can't think of a good way to do this. Is there an easy way to do this with barplot? Would some other plotter work better? I have seen things about ggplots but it seemed super complex and I wasn't sure that it would do anything better.
EDIT: If I do not choose a ylim it picks automatically and this is what it decided was best.
I disagree with it's choice.
If you don't specify ylim, R will come up with something based on the data. (Sounds like you don't like it's choice, which is fair.)
If you specify something based on the data like:
barplot(unlist(d), ylim=c(0,1.1*max(unlist(d)))
R will draw you a plot that reflects the maximum value of data. That example just takes the maximum of your values and multiplies that by 1.1 (this could be any number) to give it a little extra height. R does something similar to this when you make a scatterplot but it handles barplots slightly differently.
I have computed values for several categories for three networks. I'd like to create a bar plot in R to show the differences between these parameters for the networks. So far I plotted this with the barplot R function with the categories on the x-axis, their values on the y-axis and to each category three bars (one for each network).
But now I have one value which is much higher than all the others. Therefore the differences for the rest cannot be seen since they're represented only by a thin line because of that one large bar which almost fills the whole plot.
My idea was now to plot the values on the y-axis on an irregular scale, meaning for example, that one half represents the values from 0 to 300, and the other half from 300 to 3000. Is there any way to do this? Or a good alternative approach to handle this problem? I also thought of plotting the logarithm but unfortunatly I have also negative values.
I would suggest that an irregular scale isn't a good plan - I think it confuses viewers of the chart. Instead, you could use the layout() function to plot three separate barplots in a horizontal layout. Thus, each category could have it's own plot, with it's own scale.
If, however, you still have a single bar at 3000, while everything else is at 300, that won't really help. In that case, you could manually set your y-axis limits with ylim=c(min,max). To keep the bar from stretching off the screen, you can just use simple logic to define anything > 300 as 300, or something similar. Then, put a text point there stating the actual value (using text, maybe with arrow).
With those ideas out there, I would suggest that a graph where one value is 10x the other values might not really be worth presenting, or if it is, the main takeaway from it isn't going to be "how do values 2 and 3 compare to each other", it's going to be "holy moley look how much bigger 1 is than 2 and 3". So, it might not be a big deal if one bar is giant and two are small, as long as you aren't doing all 9 on a single plot (which would screw up other, relevant comparisons). So, if you split them using layout(), then it wouldn't be as big of a deal.
Example plot:
http://i56.tinypic.com/eagjfn.jpg
Created with:
qplot(score, ..count.., data=df, fill=method, geom='density', position='stack')
Pretty much impossible to tell what goes with what. Any way to make this better? Ideally the legend draws lines "connecting" the areas to the item in the legend. Alternatively, I'd at least need some very different filling patterns for the areas.
The human eye does not do well distinguishing between more than 7-10 different categories whether they are indicated using color, shading or pattern. Adding lines or shadings here will, I think, only make this graph harder to read.
In situations like this, I often think that it's best to take a step back and rethink what message you intend for the graph to convey. Do you really need to compare all ~23 methods in a single graph, or can the methods be placed into subgroups and compared in multiple plots or facets? Are some of the methods' curves so similar that they could be combined into a single category?
For instance, I see ~3-4 natural groups just based on the similarity of the curves in your plot. You could plot a single, representative, method from each group to illustrate the large scale differences, and then create additional plots that focus in on the differences between methods within groups.