2154-2155 / 2239
We start with the determination of the input estimates
x
1
, ...,
x
N
Each input estimate
x
i
is
obtained from a
distribution
of possible values of
the input quantity
X
.
Combined uncertainty
i
This probability distribution may be:
a)
frequency based
, that is, based on a series of N observations
X
i
,
k
of
X
i
;
∑
=
N
1
b) or it may be an
a priori distribution p
i
(x);
76
dx xpx
x
i
i
⋅
⋅
=
∫
+∞
∞−
)(
=
k
ki
i
x
N
x
1
,