where the uncertainty arises in part from the uniform and triajgular
uncertainties assumed for F and B, respectively (see text), in addition to
uncertainty is associated with the variate X associated with X defined[in Eq.
(A5). Let the subscript p on a variate denote a variate value pertaining to a
_ particular individual in the exposed population. Thus, X, =X|I{R =Rg,H =
Hp} and (X,1f ) is the sum of a presumed large number of id
independently distributed random variates. From the Lindeberg and
Limit theorems, it follows that (X, |) is approximately normally dis
with mean and variance given by
(X,1B)=ct(R, 5,18) and
oie = KRY + 73S) ~(518)']
respectively, in which
YR
=
Z;( i)
{ ;
Ry Mj
=
0.035
is the CV for uncertainty in any individual's lifetime, time-weighted average
intake based on the assumptions stated in the text and the valueslist¢d in
Table 20.
If population size N is sufficiently large to ensure tha] the
differences between sample first and second moments with respeft to
variability and their corresponding population moments are negligibje, it
follows from the definition of variability expectation that uncertainty in]
XiB
is approximately normally distributed with mean and variance givenby
(x1)) =>(x,I8) = ct(RXSIB) and
p=l
hey