Page 1 of 9

Journal for Studies in Management and Planning

Available at http://internationaljournalofresearch.org/index.php/JSMaP

e-ISSN: 2395-0463

Volume 01 Issue 05

June 2015

Available online: http://internationaljournalofresearch.org/ P a g e | 211

Moments, Characteristic and Moment Generating

Function of a Three Parameter Weibul Distribution –

an Explicit Independent Approach

Opara Jude

Department of Statistics, Imo State University PMB 2000, Owerri Nigeria

E-mail Address: judend88@yahoo.com

ABSTRACT

It is obvious that the characteristic function of three-parameter Weibull distribution can

be derived independently in an explicit form and hence the moment generating function is

then deduced from it. In this paper, the mean and variance were derived independently.

The expression of the mean, variance, skewness and kurtosis are also gotten from the

Moment Generating Function (MGF)

Key words: Moment generating function, Three-parameter Weibull

distribution, Characteristic function, Skewness, Kurtosis

THE WEIBULL DISTRIBUTION

The weibull distribution has been widely applied to many random phenomena. The

principal utility of the weibul distribution is that it affords an excellent approximation to

the probability law of many random variables (William et al; 2003). The probability

density function of three-parameter weibull distribution is given by

  exp ,

1

 

 

 

 

 x x

f x   x   ... (1)

Where,

,,  0,

are respectively the location, scale and shape parameters of the

distribution? Its cumulative distribution function is given by

 

 

  

x 

F x 1 exp ... (2)

Mean of the Weibull Distribution by definition, of a continuous random variable X is

given by;

... (3)

   



E X  xf x dx

Page 2 of 9

Journal for Studies in Management and Planning

Available at http://internationaljournalofresearch.org/index.php/JSMaP

e-ISSN: 2395-0463

Volume 01 Issue 05

June 2015

Available online: http://internationaljournalofresearch.org/ P a g e | 212

= dx x x

x

 

 

 

 

 

exp

1

dx x x x

x

 

 

 

 

 

 

exp

1

... (4)

Using integral by algebraic substitution, we have;

Let

 

  

 

x

w

x

w

1

,

1

 

 w  x   

   

1

x w

,

1

 w

dw

dx

dx w  dw

1

Substitute x and dx into equation (4) to get

   

 

 

 

  

 

w w dw w

E x w w

1

1

1

1

exp w w w  w w  dw

 

 

 

1

1

1 1

exp w w   ww dw

1

1 1

1

1

exp

  

 

  

w  wdw  wdw  

 

   

 

  exp  exp

1

w  wdw  wdw  

   

   

 

  exp  exp

1 1

1

   

 

1

1

  

     

 

1

E x 1 ... (5)

The variance of a continuous random variable is defined as;

      

2 2

var X  E X  E X ... (6)

But

   



E X  x f x dx 2 2

Page 3 of 9

Journal for Studies in Management and Planning

Available at http://internationaljournalofresearch.org/index.php/JSMaP

e-ISSN: 2395-0463

Volume 01 Issue 05

June 2015

Available online: http://internationaljournalofresearch.org/ P a g e | 213

= dx x x

x

 

 

 

 

 

exp

1

2

dx x x x

x

 

 

 

 

 

 

exp

1

2

..(7)

Let

 

x

p , 

 

 

x

p

1

,

1

 

 p  x 

  

1

 x   p

,

1

1

p

dp

dx

dx p dp

1

1

 

Substitute x and dp into equation (7) to get

   p p dp p

E x p p

1

1

1

1

2

1

2

exp

 

  

  

 

 

  

 

     p  p p p p p dp

1

2 1 1

2

1

2

2 exp  

 

   2p  p exp p dp

2

2

1

2

     

  

  

    

 

  exp p dp 2p exp p dp  p exp p dp

2

2

1

2

     

  

      

    

 

 exp p dp 2 p exp p dp  p exp p dp

1 1

2

2

1 1

1

2

   

    1

2

1

1

2

2 2

 

  

   

     

 

2

1

1

2 1

2 2 2 E x ... (8)

Substitute Equations (5) and (8) into equation (6) to get;