Normal and Other Continuous Random Variables, Study notes of Statistics

The concepts of normal (gaussian) random variables (rvs), lognormal rvs, gamma rvs, and weibull rvs. The normal distribution is defined, with mean and variance given. The empirical rule (68-95-99.7 rule) is introduced. The concept of a standard normal rv is explained, and methods for computing probabilities for normal distributions are provided. The document also covers the lognormal distribution, gamma distribution, and weibull distribution, including definitions, properties, and mean and variance calculations. Examples are given for each distribution.

Typology: Study notes

Pre 2010

Uploaded on 08/30/2009

koofers-user-6j7
koofers-user-6j7 🇺🇸

9 documents

1 / 6

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Math/Stat 370 Chapter 3 Part 2
Section 3-5 Continuous RVs
The Normal RV
Definition
a normal (Gaussian) rv X has pdf
x all for e
2
1
f(x)
)2()x(
22
where - < < + and > 0. We write X~N(,2).
N(,2) has been used to model IQ, SAT scores, heights.
Mean and Variance: E(X)= and V(X)=2
Remarks
f(x) is symmetric about .
68-95-99.7 Rule (Empirical Rule)
-68% of all values fall between .
-95% of all values fall between 2.
-99.7% of all values fall between 3.
Z~N(=0,2=1) is called a standard normal rv. We write Z~N(0,1).
z all for e
2
1
f(z)
2z
2
Probabilities for N(,2) are computed using N(0,1).
Computing N(0,1) Probabilities
Let Z~N(0,1). Values of
z2t
dte
2
1
)zZ(P)z(
2
are given in Table I on
pages 460-461.
Example 3i. Z~N(0,1)
i. P(Z 1.63) ii. P(Z > -0.59)
iii. P(-0.29 < Z < 2.11)
Computing N(,) Probabilities
3-10
pf3
pf4
pf5

Partial preview of the text

Download Normal and Other Continuous Random Variables and more Study notes Statistics in PDF only on Docsity!

Math/Stat 370 Chapter 3 Part 2

Section 3-5 Continuous RVs

The Normal RV

Definition

 a normal (Gaussian) rv X has pdf

e forall x

f(x)

(x )( 2 )

2 2  

 

where - <  < + and  > 0. We write X~N(,

2 ).

N(,

2 ) has been used to model IQ, SAT scores, heights.

Mean and Variance : E(X)= and V(X)=

2

Remarks

 f(x) is symmetric about .

 68-95-99.7 Rule (Empirical Rule)

  • 68% of all values fall between   .
  • 95% of all values fall between   2 .
  • 99.7% of all values fall between   3 .

 Z~N(=0,

2 =1) is called a standard normal rv. We write Z~N(0,1).

e forall z

f(z)

z 2

2 

Probabilities for N(,

2 ) are computed using N(0,1).

Computing N(0,1) Probabilities

Let Z~N(0,1). Values of  

z t 2 e dt

(z ) P(Z z )

2

are given in Table I on

pages 460-461.

Example 3i. Z~N(0,1)

i. P(Z  1.63) ii. P(Z > -0.59)

iii. P(-0.29 < Z < 2.11)

Computing N(,) Probabilities

Proposition

 If ~N( 0 , 1 ).

X -

X ~N(, ), then Z

2

Example 3j. X~N(3,2)

i. P(X < 1.5) ii. P(1.5 < X  3.2)

iii. Find the 95

th percentile of X.

Example 3k. X=yield strength (ksi) of steel ~ N(=42, =3.75)

i. What proportion of yield strengths falls between 30 and 50 ksi?

ii. What yield strength separates the weakest 30% from the others?

Definition

gamma function

     0

r 1 x (r ) x e dx where r > 0

Properties of()

 (k)=(k-1)! for any positive integer k

Definition

 a gamma rv X has pdf

 

0 o.w.

x e x 0

(r )

f(x)

r 1 x

r

where r,  > 0. We write X~Gam(r,).

Note:

 The density is right skewed (see plots on page 79).

 Gam(r,) models time between engine malfunctions, lengths of time to complete

maintenance checkout for an auto or aircraft engine, rainfall.

Mean and Variance E(X)=r/, V(X)=r/

2

The Weibull RV

Definition

 a Weibull rv X has pdf

 

 

 

0 o.w.

e x 0

x

f(x )

x

1

where >0 is the scale parameter and β>0 is the shape parameter. We write

X~Wei(,β).

Wei(,β) models lifetime data and describes a wide variety of failure mechanisms.

cdf of Wei(,β))

Mean and Variance

2

2 2

,V(X) 1

E( X) 1 

Example 3n. X=maximum flood level ~ Wei(=1.5, β=0.6)

a. What is the probability that flood level will be

i. more than 0.5?

ii. will not exceed 0.8?

b. Compute the mean and variance of X.