Strange precision issues in R when computing cumulative binomial probability

Question

I\'ve been running into some weird problems when using this code:

positions<-c(58256)
occurrencies<-c(30)
frequency<-c(11/5531777)
length<-c(4)

         


        

Answer 1

You can avoid your for loop by doing

prob<-0
i    <- 0:(occurrencies-1)
pow  <- frequency^i
pow1 <- (1-frequency)^(positions-i)
bin  <- choose(positions, i)
prob <- cumsum(prob+(bin*pow*pow1))
[1] 0.8906152 0.9937867 0.9997624 0.9999932 0.9999998 1.0000000 1.0000000 1.0000000 1.0000000
[10] 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000
[19] 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000 1.0000000
[28] 1.0000000 1.0000000 1.0000000

I don't know if this is your desired result, but surely you can avoid the for loop going this fashion.

See @Ben Bolker's comment and take a look at pbinom function.

Answer 2

Following Ben Bolker's advice to see ?pbinom

pbinom(q = occurencies, size = positions, prob = frequency, lower.tail = FALSE)