Kernel density estimators in Bishop book unclear formulas

18 Views Asked by At

In Bishop's book (Pattern recognition and machine learning, pag 122) there is an unclear passage for me in deriving certain formulas:

$E[K/N] = P$ and $var[K/N] = P(1-P)$

Considering binomial probability

$P(x=x) = \binom{N}{x} P^{x} \cdot (1 - P)^{(N-x)}$

I know that the expected value and variance is given by:

$E(X) = N \cdot x$

$var(X) = N \cdot x \cdot (1-x)$

As a first thing since Bishop refers to the fraction of points that fall in the bin, i.e., K/N, I would be tempted to substitute N for K/N, however, there are some steps (perhaps algebraic) that I am missing.

enter image description here