Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

Cumulative Plot with Given X-Axis

Tags:

r

statistics

I have data that looks like this. In which I want to plot accumulative value of dat1 with respect to x-axis. Also plot it together with dat2.

#x-axis dat1              dat2
-10     0.0140149       0.0140146
-9      0.00890835      0.00891768
-8      0.00672276      0.00672488
-7      0.00876399      0.00879401
-6      0.00806879      0.00808141
-5      0.0088366       0.00885121
-4      0.00856872      0.00857769
-3      0.0195384       0.0196094
-2      0.0160239       0.0161829
-1      0.0254455       0.0257845
0       0.0397413       0.0400913
1       0.0743316       0.0755453
2       0.0247501       0.0253324
3       0.0214285       0.021778
4       0.0241462       0.0244967
5       0.0150943       0.015241
6       0.0141398       0.0142373
7       0.0101515       0.0102948
8       0.0308843       0.031294
9       0.0095504       0.00960626
10      0.00729676      0.0073713

What's the common way to do it in R?

I looked at ECDF from Hmisc, it doesn't seem to do what I want. In particular it doesn't allow us to give x-axis value.

like image 939
neversaint Avatar asked Mar 18 '09 08:03

neversaint


People also ask

What is x-axis of cumulative distribution function?

PetroWiki. A graph the horizontal axis of which is a variable X and the vertical axis of which ranges from 0 to 1. There are two types: ascending and descending. In an ascending CDF, a point (x, y) indicates that the probability that X is less than or equal to x is y.

How do you plot cumulative data in R?

Plot cumulative distribution function in base R To plot a CDF function in base R, we first calculate the CDF by using the ecdf() function. Then we use the plot() function to plot the CDF plot in the R Language. The plot function takes the result of the ecdf() function as an argument to plot the CDF plot.


2 Answers

I think the function you are looking for is cumsum() which will do a cumulative sum on a vector.

#put your data into 3 vectors
x<-c(-10,-9,-8,-7,-6,-5,-4,-3,-2,-1,0,1,2,3,4,5,6,7,8,9,10)
dat1<-c(0.0140149,0.00890835,0.00672276,0.00876399,0.00806879,0.0088366,0.00856872,0.0195384,0.0160239,0.0254455,0.0397413,0.0743316,0.0247501,0.0214285,0.0241462,0.0150943,0.0141398,0.0101515,0.0308843,0.0095504,0.00729676)
dat2<-c(0.014015,0.008918,0.006725,0.008794,0.008081,0.008851,0.008578,0.019609,0.016183,0.025785,0.040091,0.075545,0.025332,0.021778,0.024497,0.015241,0.014237,0.010295,0.031294,0.009606,0.007371)

#create a new vector called cdat1 to hold the cumulative sum
cdat1<-cumsum(dat1)
plot(x,cdat1)
points(x,dat2,col="red")

I use the function points above in order to add dat2 to the existing plot. Run this in R and see if it gives you what you need.

image of plot

like image 118
JD Long Avatar answered Oct 21 '22 08:10

JD Long


If you have that data in a text file (e.g. data.txt) you can also do the following:

A <- read.table("data.txt",header=TRUE)
attach(A)
plot(x.axis, cumsum(dat1))
points(x.axis, cumsum(dat2), col='red')

As JD Long said, the cumsum function is what you were looking for.

like image 24
Christopher DuBois Avatar answered Oct 21 '22 06:10

Christopher DuBois