You are on page 1of 20

Distributions

1-20

FRMFinancial Risk Manager

Continuous Uniform Distribution

Uniform [a, b] Distribution

f(x)

probability density function f(x) = 1/(b a)

The area under f(x) from x1 to x2 = P(x1X x2)


= (x2 x1)/(b a)
a

x2

x1

a+b
E(X) =
2
2-20

(b - a) 2
Var(X)
2

FRMFinancial Risk Manager

Example
What is the probability of an outcome being between 15 and 25 for a random
variable that follows a continuous uniform distribution over the range of 12 to
28?

A. 0.509.
B. 0.625.
C. 1.000.

D. 1.600.

AnswerB

3-20

25 15

P 15 X 25
0.625
28 12

FRMFinancial Risk Manager

Bernoulli Distribution and Binomial Distribution

Bernoulli random variable


P(Y=1)=p P(Y=0)=1-p
Binomial random variable
the probability of x successes in n trails

p(x) = P(X = x) = Cnx px (1-p)n-x


Expectations and variances

4-20

Expectation

Variance

Bernoulli random variable (Y)

p(1-p)

Binomial random variable (X)

np

np(1-p)

FRMFinancial Risk Manager


A recent study indicated that 60% of all businesses have a fax machine.
Assuming a binomial probability distribution, what is the probability that exactly
four businesses will have a fax machine in a random selection of six businesses?

A. 0.138.
B. 0.276.
C. 0.311.

D. 0.324.

AnswerC

P(X = 4) = C 60% (1-60%)6-4 0.311


4
6

5-20

FRMFinancial Risk Manager

Poisson Distribution
Poisson random variable X refers to the number of successes per unit, the
parameter lambda () refers to the average or expected number of successes per
unit.

x e-
P(X = x) =
x!

An interesting feature of the Poisson distribution is that both its mean and
variance are equal to the parameter,

6-20

FRMFinancial Risk Manager

Answer: D

=3 2=6
x e- 65 e6
P(X = 5) =

16.06%
x!
5!

7-20

FRMFinancial Risk Manager

Normal Distribution
The shape of the density function
The normal curve is symmetrical
The two halves are identical

Theoretically, the
curve extends to -

Properties:

Theoretically, the
curve extends to +

The mean, median, and mode are


equal.

X ~ N(, )
Symmetrical distribution: skewness=0; kurtosis=3
A linear combination of normally distributed random variables is also
normally distributed.
8-20

FRMFinancial Risk Manager

Confidence Intervals for a Normal Distribution

The Confidence Intervals

X ~ N( , )

-2.58
-1.65
-1.96

+1.65 +2.58
+1.96

90%

95%
99%
9-20

FRMFinancial Risk Manager

10-20

FRMFinancial Risk Manager

Standard Normal Distribution

Standard normal distribution


N(0,1) or Z
Standardization: if X~N (, ),
then Z X ~ N(0,1)

Z-table
Example:
X ~ N (709) , compute the probability of X 75.9

First Z 75.9 70 1.96 , then compute P(Z 1.96) 1 0.975 0.025


3

11-20

FRMFinancial Risk Manager

Z-Table

12-20

FRMFinancial Risk Manager

Answer: A

13-20

FRMFinancial Risk Manager

Lognormal Distribution

If lnX is normal, then X is lognormal


Right skewed
Bounded from below by zero
Lognormal distribution is used to model asset prices

14-20

FRMFinancial Risk Manager

10

15-20

FRMFinancial Risk Manager

Central Limit Theorem

For simple random samples of size n from a population with a mean


and a variance , the sampling distribution of the sample mean

approaches N(, /n) if the sample size is sufficiently large (n


30).

X N ( ,

16-20

FRMFinancial Risk Manager

t-distribution

Symmetrical
Degrees of freedom (df): n-1
Less peaked than a normal distribution (fatter tails)
As the degrees of freedom gets larger, the shape of t-distribution
approaches standard normal distribution

X-u
~ t(n-1)
s

17-20

FRMFinancial Risk Manager

Chi-Square Distribution
The chi-square test statistic, , with n-1 degrees of freedom, is computed as:
2

n21

(n 1) s 2

02

df=30

18-20

FRMFinancial Risk Manager

F-Distribution
The test statistic for the F-test is the ratio of the sample variances

s12
F= 2 ~ F(n1 -1),
s2

(n 2 -1)

df1 =10, df 2 =10

19-20

FRMFinancial Risk Manager

FRM

20-20

FRMFinancial Risk Manager