PMF and CDF from a piecewise function in R

Question

Given

 p<-function(x){if ( x==-4) {
    0.1
  } else if (x==-1 ) {
   0.2
  } else if ( x==1) {
    0.6
  } else if (x==2 ) {
    0.1
  } else {
   0
  }

How would I plot the PMF and CDF. No distribution is specified so I cant use the build in commands. What would be the best way to plot these

Lennart Oelschläger · Answer 1 · 2021-02-18T13:29:24.730

0

Easy and simple solution:

### support
x = seq(-4,2,1)
### PDF
plot(x,sapply(x,p),ylab="PDF")
### CDF
plot(x,cumsum(sapply(x,p)),ylab="CDF")

edited Feb 18 '21 at 13:29

answered Feb 18 '21 at 12:30

1 question though. Why do we not write all the x's in the seq? i.e (-4, -1, 0, 1, 2) – Feb 18 '21 at 12:31
You can do that, just specify `x = c(-4, -1, 0, 1, 2)`. – Lennart Oelschläger Feb 18 '21 at 13:31

ThomasIsCoding · Accepted Answer · 2021-02-18T12:43:57.597

0

Try this for PMF

df <- data.frame(x = c(-4, -1, 1, 2), p = c(.1, .2, .6, .1))
plot(df, type = "l")
points(df)

For CDF, you can try a similar one

plot(df_cdf <- transform(df, p = cumsum(p)), type = "l")
points(df_cdf)

edited Feb 18 '21 at 12:43

answered Feb 18 '21 at 12:35

ThomasIsCoding

Is there any way to connect the dots? or make a curve. – Feb 18 '21 at 12:35
@mathstudent23 Use `type = "l"` – ThomasIsCoding Feb 18 '21 at 12:39
Okay would a cdf be possible with this method? And would you be able to draw "lines" from the given x up to its probability? – Feb 18 '21 at 12:41

2 Answers2