Next: 12.3 Regression Models
Up: 12. Computational Methods in
Previous: 12.1 Introduction
Let us now consider the class of the lifetime distributions,
whose distribution functions are expressed by
|
(12.11) |
where is also a distribution function. For the
Weibull model,
is an exponential
distribution. Nagatsuka and Kamakura
([24], [25]) proposed a new method using
the location-scale-free
transformation of
data set to estimate the power parameter in the Castillo-Hadi model ([2]). That is, let
be
independently distributed according to the distribution
function (12.11). Consider the
-transformation to be defined as
|
(12.12) |
where is the -th order statistic of 's. The
new random variables 's derived by this -transformation
are then free from location and scale parameter. The
arithmetic mean of 's gives the approximation to the
original distribution of . Let
be
i.i.d. distributed with common distribution function ,
and let the -th order statistic have the marginal
distribution function
. Then
|
(12.13) |
This equation indicates that the arithmetic mean of the
marginal distributions of order statistics is exactly the
original distribution. In the case of the Castillo-Hadi Model,
[25] provided a theorem regarding this
approximation, i.e.,
Theorem 1 ([
25])
The mixture of the marginal distributions of ,
:
|
(12.14) |
is the approximate distribution of 's and the limiting
distribution (12.14) is the power function
distribution with parameter . That is
In the case of the Weibull distribution, the marginal
distribution of is calculated as
where
Calculations show that
has a first moment of
where
is the incomplete
generalized gamma function defined by
Now, an estimating of the shape parameter is obtained by
equating the theoretical population mean with sample mean of
-transformed 's. [24] provided a table for
obtaining estimates and concluded based on simulation studies
that the robust estimate of is possible without using any
existing threshold parameter.
Next: 12.3 Regression Models
Up: 12. Computational Methods in
Previous: 12.1 Introduction