3 TRUNCATED POISSON MIXTURE
The Poisson distribution is a discrete probability distribution. The Poisson distri-
bution is some times truncated, i.e. the random variables are assigned numbers that
are greater than zero. The Poisson distribution is a discrete distribution used for the
interval counts of events that randomly occur in given interval (or space)[3]. The
probability mass function (pmf) is
P (N = n) =
λ
n
e
−λ
n!
, n = 0, 1, 2, 3...; λ > 0.
with expectation E(N) = λ and variance V (N) = λ. The probability generating
function of the Poisson distribution is G(t) = e
λ(t−1)
and the moomemt generating
function (mgf) is M(t) = e
λ(e
t
−1)
, where the events occur on a given time t.
The truncated Poisson is a discrete probability distribution which is used to de-
scribe events that occur per unit time and can not be a zero event. In this case,
the starting point will not be zero but 1. This process is termed as the truncated
Poisson distribution or the zero truncated Poisson distribution. The pmf of the zero
truncated Poisson is given below as
P (N = n) =
λ
n
e
−λ
(1 − e
−λ
)n!
, n = 1, 2 . . . .
with an expectation of E(N ) =
λ
1−e
−λ
and a variance of V (N) =
λ
(1−e
−λ
)
2
.
If the random variable X
i
follows a continuous probability distribution and Z|N =
min(X
1
, ...., X
n
), then we can find a distribution for the first other statistic X
(1)
when
the sample size is fixed or random. In the next section, the paper will focus more on a
general formula for finding the cdf and pdf of a random variable with any continuous
probability distribution like uniform, exponential, etc. and a random sample size(N).
9