Introduction to Probability and Statistics for Engineers and Scientists

4.5Properties of the Expected Value 111

-1

p (-1) = .10,

center of gravity = .9

0

p (0) = .25,

1

p (1) = .30,

2

p (2) = .35

FIGURE 4.4

it is natural to define the expected value ofXby

E[X]=

∫∞

−∞

xf(x)dx

EXAMPLE 4.4d Suppose that you are expecting a message at some time past 5P.M. From
experience you know thatX, the number of hours after 5P.M. until the message arrives,
is a random variable with the following probability density function:

f(x)=







1

1.5

if 0<x<1.5

0 otherwise

The expected amount of time past 5P.M.until the message arrives is given by

E[X]=

∫1.5

0

x

1.5

dx=.75

Hence, on average, you would have to wait three-fourths of an hour. ■

REMARKS

(a) The concept of expectation is analogous to the physical concept of the center of gravity
of a distribution of mass. Consider a discrete random variableXhaving probability mass
functionP(xi),i≥1. If we now imagine a weightless rod in which weights with mass
P(xi),i≥1 are located at the pointsxi,i≥1 (see Figure 4.4), then the point at which
the rod would be in balance is known as the center of gravity. For those readers acquainted
with elementary statics, it is now a simple matter to show that this point is atE[X].*
(b)E[X]has the same units of measurement as doesX.

4.5Properties of the Expected Value

Suppose now that we are given a random variableXand its probability distribution (that
is, its probability mass function in the discrete case or its probability density function in
the continuous case). Suppose also that we are interested in calculating, not the expected

• To prove this, we must show that the sum of the torques tending to turn the point aroundE[X]is equal to 0. That
  is, we must show that 0=∑i(xi−E[X])p(xi), which is immediate.