The values of FX[x(i)
X. For example, with the aid of Table A.3 for standardized normal random
variable U, we have
and so on.
In order to determine d 2 , it is constructive to plot F^0 (x) and FX(x) as
functions of x, as shown in Figure 10.2. It is clearly seen from the figure that
the maximum of the differences between F^0 (x) and FX(x) occurs at x x(4)
With 0 05 and n 10, Table A.6 gives
F^0 (x) F
X(x)
1.0
0.8
0.6
0.4
0.2
0
27 29 31 33
x
(^0) F
(x
),
FX
(x
)
Figure 10. 2^0 F(x) and FX (x) in Example 10.5.
Model Verification 329
] cannowbefoundbased on distribution N(30 3, 3 14) for::
FX 27 : 4 FU
27 : 4 30 : 3
3 : 14
p
FU 1 : 64
1 FU 1 : 64 1 0 : 9495 0 : 0505 ;
FX 28 : 7 FU
28 : 7 30 : 3
3 : 14
p
FU 0 : 90
1 0 : 8159 0 : 1841 ;
d 2 jF^0
29 : 1 FX
29 : 1 j 0 : 4 0 : 2483 0 : 1517 :
:
c 10 ; 0 : 05 0 : 41 :
29 1. H ence,: