You are on page 1of 16

Chapter 6 Continuous Random Variable

6.1 Continuous Probability Distribution



A continuous random variable is a random variable whose values are not countable, it can
assume any value over an interval or intervals.

For example, the life of a battery, heights of persons, time taken to complete a test, etc, are
continuous random variables.






The probability that a continuous random variable X assumes a single valueis always zero.
That is, 0 ) ( = = a X P


Therefore, for a continuous random variable X, we will also call its probability distribution
curve, probability density function. This function, ) (x f , satisfies the following properties:

i) 0 ) ( > x f , for all x.
ii) The total area between the curve of ) (x f and the x-axis is always 1, that is
1 ) ( =
}


dx x f


iii) dx x f b X a P
b
a
}
= < < ) ( ) ( .
Note that : ) ( ) ( ) ( ) ( b X a P b X a P b X a P b X a P s s = s < = < s = < <




























Mean
Let X be a continuous random variable with probability density function ) (x f . The mean or
expected value of X is
}


= = dx x f x X E ) ( ) ( .

Variance
Let X be a continuous random variable with probability density function ) (x f and mean .
The variance of X is
}


= = = dx x f x X E X Var
X
) ( ) ( ] ) [( ) (
2 2 2
o
or
2 2 2 2 2
) ( ) ( ) ( o = = =
}


dx x f x X E X Var
X


Example 6.1: The continuous random variable of X has the probability density function as
follows:

/ 8 0 4
( )
0
x x
f x
otherwise
< <
=


Find E(X) and Var(X).

Solution:































6.2 The Uniform Distribution

A continuous random variable, X has a uniform distribution if its values spread evenly over
the range of [a,b] and denoted as X~Uniform(a,b). Thus the probability density function f (x)
is
,
1
) (
a b
x f

= a s x s b



f(x)
f(x) = 1/(b a)

1/(b a)


x = a x = b x

The graph of a uniform distribution results in a rectangular shape.


The Uniform probability for a certain range
Unifrom Probability random variable, X between two potins x
1
and x
2
can be obtained by
integrating the density function from x
1
to x
2
, as follows:

}
= s s
2
1
) ( ) (
2 1
x
x
dx x f x X x P

}

=
2
1
1
x
x
dx
a b

) (
1
1 2
x x
a b

=
The output of the integration above is as same as area for a rectangle with the width of (x
2

x
1
) and the height of 1/(b - a). Therefore, without requiring any integrating action, the uniform
probability X by these points x
1
and x
2
in the range of [a, b] is:


f(x) = 1/(b a)
1/(b a)



a x
1
x
2
b x


Mean X dan variance X are given as follows:

E(X) =
x
= (a + b)/2

Var(X) = o
x
2
= (b a)
2
/12




Example 6.2: Let the random variable X is distributed as continuous uniform in the range of 2 to 6.
Find
(i) Probability density X (ii) P(3 s X s 4)
(iii) P(X > 3.5) (iv) E(X) and Var(X)

Solutions:





















Example 6.3: The time when a used car is experiencing a first broken damage after being sold, is
known distributed as continuously uniform in range of 0 to 6 months. Assumed a dealer guarantees
that the car is not experiencing any broken damage within 60 days period after being sold,
(i) What is the probability that the used car that has been sold will be firstly broken in the period of
the warranty?
(ii) When will be the sold car expected to suffer the first broken?

Solution:



















6.3 The Normal Distribution

The normal distribution is one of the many probability distributions that a continuous random
variable can possess. The normal distribution is the most important and most widely used of
all probability distributions.

The normal probability distribution or the normal curve is a bell-shaped (symmetric) curve,
its mean is denoted by and its standard deviation by .

The properties of a normally distributed variable:
1. The total area beyond this curve, f(x) is 1.0.


Area = 1.0 f(x)

x


2. The curve f(x) is symmetric around its mean, . Therefore:
Area of left side of = Area of right side of = 0.5



0.5 0.5 x



3. Since f(x) is symmetric towards its mean, all values of mean, median and mode are all
same.



x
min = median = mod


4. The tail at the left and right sides of the curve f(x) extend to the infinity. The curve does not
touch the x -axis; but it approximates the x -axis until the area between the points x =
3o and x = + 3o is approaching 1.0.

Area ~ 1.0


x x +
3o + 3o







The probability density function of a normal distribution is
( )
0 , ,
2
exp
2
1
) (
2
2
> < < < <
(


= o
o

t o
x
x
x f


2
2
2
) (
2
1
) (
o

t o

=
x
e x f


x

If X~ ) , (
2
o N , then the mean and the variance of X are
= = ) (X E
X

and
2 2
) ( o o = = X V
X
, respectively.

Three normal distributions with the same value of mean, i.e.
1
=
2
=
3
= 20 and the
respectively standard deviations o
1
= 5, o
2
= 10,

o
3
= 16.





Three normal distributions that contain the respective means
1
= 10,
2
= 20,
3
= 30 and the
same standard deviation, o
1
= o
2
= o
3
= 16.







6.3.1 The Standard Normal Distribution

A normally distributed variable having mean 0 and standard deviation 1 is said to have the
standard normal distribution.

Standard Normal Probability Density Function:
,
2
1
) (
2
2
z
e z f

=
t
< < z


f(z)




-3 -2 -1 = 0 1 2 3 z



To calculate the standard Normal probability between two points

and

.

}
= s s
2
1
) ( ) (
2 1
z
z
dz z f z Z z P
dz e
z
z
z
}

=
2
1
2
2
2
1
t


The above integral cannot be solved explicitly. However the probability can be easily
determined using Standard Normal Table which reports the probability that a standard normal
random variable has probability less than or equal to constant z.

Assumed

and

are two values of standard normal distribution, with

<

. The
following are true towards the probability of Z:


P(

s Z s

) area of region between


= P(Z s

P(Z s

and


z


P(Z s

) = 1 P(Z >

) area of region from left of

z


P(Z >

) = 1 P(Z s

) area of region from right of

z

The location points of

and

in the z-axis depend at their sign; if the sign is positive, the


location point is at the right of the zero point and if the sign is negative, the location of the
point is at the left from zero point.
Finding Probabilities When Given z Values
Table 4. The Normal Distribution Function (Cambridge Statistical Tables) lists the
probability of standard normal distribution at the range of ( , z), i.e. the probability Z in
form of:
) ( ) ( z Z P z s = u
This probability is equal to the area of region beyond the normal curve, from - to z, i.e. the
area of the region in the left hand side of z.


u(z) N(0,1)
0 z

The following is the example of the values of z and its probability, ) ( ) ( z Z P z s = u , which can
be obtained from the standard normal Table, Table 4: The Normal Distribution Function :

z u(z) z u(z)
0.00 0.5000 0.40 0.6554
.10 .5398 .50 .6915 ) 10 . 1 ( ) 10 . 1 ( s = Z P u
.20 .5793 1.10 .8643 = 0.8643
.30 .6179 1.72 .9573

Example 6.4: Find the following probability Z and sketch the probability area
(i) P(Z s 0.00) (ii) P(Z s 1.12) (iii) P(Z > 1.72) (iv) P(0.30 s Z s 1.72)

Solutions:























If z is a negative value, the probability of Z is determine using the symmetry property of the
standard Normal distribution about their mean, 0 by assuming z is a positive value where z > 0.

P(Z s z) = P(Z > z)
= 1 P(Z s z)

- z 0 z

and P(Z > z) = P(Z s z)


- z 0 z - z 0 z


Example 6.5: Find the probabilities of the following Z. Shade the areas for each of the
probabilities.
(i) P(Z s 1.10) (ii) (Z > 1.10) (iii) P( Z s 1.10) (iv) P( Z > 1.10)
Solution:




























6.4 Standardizing a Normal distribution
The standardized version of a normally distributed variable X ,
o

=
x
Z , has the standard
normal distribution. This is also called the z values or z scores.

Example 6.6: Let x be a random variable that has a normal distribution with a mean of 40 and
a standard deviation of 5. Find the following probabilities for this normal distribution.
a) P(X > 55) b) P(X < 49) c) P(42 < X < 50) d) P(32 < X < 38)

Solution:











Example 6.7: Hupper Corporation produces many types of soft drinks, including Orange
Cola. The filling machines are adjusted to pour 12 ounces of soda can of Orange Cola.
However, the actual amount of soda poured into each can is not exactly 12 ounces; it varies
from can to can. It has been observed that the net amount of soda in such a can has a normal
distribution with a mean of 12 ounces and a standard deviation of 0.015 ounce.
a) What is the probability that a randomly selected can of Orange Cola contains 11.97 to
11.99 ounces of soda?
b) What percentage of the Orange Cola cans contain 12.02 to 12.07 ounces of soda?

Solution:


















Example 6.8: The life span of a calculator has a normal distribution with a mean of 54
months and a standard deviation of 8 months. The company guarantees that any calculator
that starts malfunctioning within 36 months of the purchase will be replaced by a new one.
About what percentage of calculators made by this company are expected to be replaced?

Solution:


























6.5 Determining the z and x values when an area under the Normal
distribution curve is known

Sometimes we have to determine the interval under the normal curve given a fixed probability. This
situation is opposite of what we encounter previously.
To determine the values of z using a fixed probability, we still use Table 4 but now we read the table
from right (column (z)) to left, (column z).

z u(z) z u(z)
0.00 0.5000 0.40 0.6554
.10 .5398 .50 .6915 ) 10 . 1 ( ) 10 . 1 ( 8643 . 0 s = u = Z P
.20 .5793 1.10 .8643
.30 .6179 1.72 .9573

Example 6.9.: Find the values of z so that
(i) P(Z s z) = 0.9573 (ii) P(Z > z) = 0.3085
(iii) P(-z s Z s z) = 0.50 (iv) P(Z s z) = 0.10
Solution:

















Example 6.10: The life span of a calculator has a normal distribution with a mean of 54
months and a standard deviation of 8 months. What should the warranty period be to replace a
malfunctioning calculator if the company does not want to replace more than 1% of all the
calculators sold?

Solution:

















6.6 The Normal Approximation to the Binomial Distribution

- The normal distribution can be used as an approximation to the binomial distribution when
np and nq are both greater than 5, that is if 5 5 > > nq and np . That is,
if X ~ B(n, p), then X ~ N(np, npq)

- An addition of 0.5 and/or subtraction of 0.5 from the value(s) of x when the normal
distribution is used as an approximation to the binomial distribution is called the continuity
correction factor.
Example 6.11: According to an estimate, 50% of the people in America have at least one
credit card. If a random sample of 30 persons is selected, what is the probability that 19 of
them will have at least one credit card?

Solution:
Method 1: exact answer
) 5 . 0 , 30 ( ~ = = p n B X ,
using the Binomial distribution, P(X = 19) 0509 . 0 ) 5 . 0 ( ) 5 . 0 (
11 19
19
30
= = C

Method 2: approximation
Mean 15 ) 5 . 0 ( 30 = = = np
Variance 5 . 7 ) 5 . 0 )( 5 . 0 ( 30
2
= = = npq o




) 5 . 19 5 . 18 ( ) 19 ( < < = = X P X P

|
|
.
|

\
|
< <

=
5 . 7
15 5 . 19
5 . 7
15 5 . 18
Z P
) 643 . 1 278 . 1 ( < < = Z P
0498 . 0 =



Example 6.12: According to a joint reader survey by USATODAY.com and
TripAdvisor.com, 34% of the people surveyed said that the first thing they do after checking
into a hotel is to adjust the thermostat (USA TODAY, June 12, 2009). Suppose that this result
is true for the current population of all adult Americans who stay in hotels. What is the
probability that in a random sample of 400 adult American who stay in hotels, 115 to 130 will
say that the first thing they do after checking into a hotel is to adjust the thermostat?

Solution:

















Example 6.13: According to an American Laser Centers survey, 32% of adult men said that
their stomach is the least favorite part of their body (USA TODAY, March 10, 2009). Assume
that this percentage is true for the current population of adult men. What is the probability that
170 or more adult men in a random sample of 500 will say that their stomach is the least
favorite part of their body?

Solution: