Estimators for the mean and standard deviation of the difference and sum of the
statistics S
and S
1
In these expressions, X
samples taken from the two populations, and
of the populations of the statistics S
taken.
Confidence intervals for sums and differences of mean values
If the population variances
the difference and sum of the mean values of the populations, i.e.,
given by:
⎛
⎜
(
X
X
⎜
1
2
⎝
For large samples, i.e., n
population variances
sum of the mean values of the populations, i.e.,
⎛
⎜
(
X
X
⎜
1
⎝
If one of the samples is small, i.e., n
equal, population variances
the variance of
are given by:
2
ˆ
X
X
S
S
1
1
2
and X
1
2
and
1
2
1
)
z
2 /
n
1
> 30 and n
1
2
2
=
, the confidence intervals for the difference and
1
2
2
S
1
)
z
2
2 /
n
1
2
1
2
, as s
= [(n
1
2
p
ˆ
,
2
S
S
1
2
are the values of the statistics S
2
S1
and S
from which the samples were
1
2
2
are known, the confidence intervals for
2
2
2
( ,
X
X
1
n
2
> 30, and unknown, but equal,
2
2
S
2
( ,
X
X
1
n
2
< 30 or n
1
2
2
=
, we can obtain a "pooled" estimate of
2
2
-1) s
+(n
-1) s
1
1
2
2
2
S
1
S
2
n
n
1
2
1
2
2
and
are the variances
S2
1
)
z
2
2 /
n
1
, are given by:
1
2
2
S
1
)
z
2
2 /
n
1
< 30, and with unknown, but
2
]/( n
+n
-2).
2
1
2
and S
from
2
, are
1
2
⎞
2
2
⎟
2
⎟
n
⎠
2
⎞
2
S
⎟
2
.
⎟
n
⎠
2
Page 18-26