3

I am having hard time to understand how the pnbinom(q, size, prob, mu, lower.tail = TRUE, log.p = FALSE) in R to the scipy.stats.nbinom.pmf(k, n, p, loc=0) in SciPy.

For the R function, the definitions of the parameters are as follows.

q   =vector of quantiles.

size    = target for number of successful trials, or dispersion parameter (the shape parameter of the gamma mixing distribution). Must be strictly positive, need not be integer.

prob    =probability of success in each trial. 0 < prob <= 1.

mu  = alternative parametrization via mean: see ‘Details’.

log, log.p  =logical; if TRUE, probabilities p are given as log(p).

lower.tail =    logical; if TRUE (default), probabilities are P[X ≤ x], otherwise, P[X > x].

For the SciPy function, the parameters are defined as follows.

 n is the    number of successes
 p is the probability of a single success.

For example, if 

k=20

a=1.2

p=0.1

In R, pnbinom(k,a,p) = 0.8518848. Here, k is plugged into q i.e. the vector of quantiles, a is plugged into size, and p is plugged into 'prob'.

On the other hand, in SciPy, I assumed n is what used as size and p is what we used as prob in R. In that setting, nbinom.pmf(k, a, p) = 0.01530062999480606.

Could anyone please help to identify what I am missing?

Omar Shehab
  • 1,004
  • 3
  • 14
  • 26

1 Answers1

4

nbinom.pmf(k, a, p) returns the pmf (probability mass function), while pnbinom(k, a, p) is the cdf (cumulative distribution function).

Try nbinom.cdf(k, a, p) to get the cdf from scipy, or dnbinom(k, a, p) to get the pmf in R.

RyanFrost
  • 1,400
  • 7
  • 17