4-1
Mean and Variance of the Normal Distribution (CD Only)
In the derivations below, the mean and variance of a normal random variable are shown
to be and ^2 , respectively. The mean of xis
By making the change of variable , the integral becomes
The first integral in the expression above equals 1 because is a probability density
function and the second integral is found to be 0 by either formally making the change of vari-
able uy^2 2 or noticing the symmetry of the integrand about y0. Therefore, E(X) .
The variance of Xis
By making the change of variable , the integral becomes
Upon integrating by parts with and V(X) is found to be.
4-8 CONTINUITY CORRECTIONS TO IMPROVE
THE APPROXIMATION
From Fig. 4-19 it can be seen that a probability such as P(3 X 7) is better approximated
by the area under the normal curve from 2.5 to 7.5. This observation provides a method to im-
prove the approximation of binomial probabilities. Because a continuous normal distribution
is used to approximate a discrete binomial distribution, the modification is referred to as a
continuity correction.
dvy ^2
ey
(^2) 2
22
uy dy,
V 1 X 2 ^2
y^2
ey
(^2) 2
22
dy
y 1 x (^2)
V 1 X 2
1 x 22
e^1 x^2
(^2) 2 2
22
dx
ey
(^2) 2
22
E 1 X 2
ey
(^2) 2
22
dy
y
ey
(^2) 2
22
dy
y 1 x (^2)
E 1 X 2
x
e^1 x^2
(^2) 2 2
22
dx
If Xis a binomial random variable with parameters nand p, and if x 0, 1, 2,p , n,
the continuity correctionto improve approximations obtained from the normal dis-
tribution is
and
P 1 xX 2 P 1 x0.5X 2 P °
x0.5np
2 np 11 p 2
Z¢
P 1 Xx 2 P 1 Xx 0.5 2 P °Z
x 0.5np
2 np 11 p 2
¢
PQ220 6234F.CD(04) 5/13/02 11:55 M Page 1 RK UL 6 RK UL 6:Desktop Folder:TEMP WORK:MONTGOMERY:REVISES UPLO D CH114 FIN L:Quark