# Logistic Confidence Bounds

Jump to navigation Jump to search
 Chapter 14.1: Logistic Confidence Bounds

 Chapter 14.1 Logistic Confidence Bounds

Available Software:
Weibull++

More Resources:
Weibull++ Examples Collection

Download Reference Book:
Life Data Analysis (*.pdf)

## Confidence Bounds

In this section, we present the methods used in the application to estimate the different types of confidence bounds for logistically distributed data. The complete derivations were presented in detail (for a general function) in Confidence Bounds.

### Bounds on the Parameters

The lower and upper bounds on the location parameter ${\displaystyle {\widehat {\mu }}\,\!}$ are estimated from

${\displaystyle {{\mu }_{U}}={\widehat {\mu }}+{{K}_{\alpha }}{\sqrt {Var({\widehat {\mu }}){\text{ }}}}{\text{ (upper bound)}}\,\!}$
${\displaystyle {{\mu }_{L}}={\widehat {\mu }}-{{K}_{\alpha }}{\sqrt {Var({\widehat {\mu }}){\text{ }}}}{\text{ (lower bound)}}\,\!}$

The lower and upper bounds on the scale parameter ${\displaystyle {\widehat {\sigma }}\,\!}$ are estimated from:

${\displaystyle {{\sigma }_{U}}={\widehat {\sigma }}{{e}^{\tfrac {{{K}_{\alpha }}{\sqrt {Var({\widehat {\sigma }}){\text{ }}}}}{\widehat {\sigma }}}}({\text{upper bound}})\,\!}$

${\displaystyle {{\sigma }_{L}}={\widehat {\sigma }}{{e}^{\tfrac {-{{K}_{\alpha }}{\sqrt {Var({\widehat {\sigma }}){\text{ }}}}}{\widehat {\sigma }}}}{\text{ (lower bound)}}\,\!}$

where ${\displaystyle {{K}_{\alpha }}\,\!}$ is defined by:

${\displaystyle \alpha ={\frac {1}{\sqrt {2\pi }}}\int _{{K}_{\alpha }}^{\infty }{{e}^{-{\tfrac {{t}^{2}}{2}}}}dt=1-\Phi ({{K}_{\alpha }})\,\!}$

If ${\displaystyle \delta \,\!}$ is the confidence level, then ${\displaystyle \alpha ={\tfrac {1-\delta }{2}}\,\!}$ for the two-sided bounds, and ${\displaystyle \alpha =1-\delta \,\!}$ for the one-sided bounds.

The variances and covariances of ${\displaystyle {\widehat {\mu }}\,\!}$ and ${\displaystyle {\widehat {\sigma }}\,\!}$ are estimated from the Fisher matrix, as follows:

${\displaystyle \left({\begin{matrix}{\widehat {Var}}\left({\widehat {\mu }}\right)&{\widehat {Cov}}\left({\widehat {\mu }},{\widehat {\sigma }}\right)\\{\widehat {Cov}}\left({\widehat {\mu }},{\widehat {\sigma }}\right)&{\widehat {Var}}\left({\widehat {\sigma }}\right)\\\end{matrix}}\right)=\left({\begin{matrix}-{\tfrac {{{\partial }^{2}}\Lambda }{\partial {{\mu }^{2}}}}&-{\tfrac {{{\partial }^{2}}\Lambda }{\partial \mu \partial \sigma }}\\{}&{}\\-{\tfrac {{{\partial }^{2}}\Lambda }{\partial \mu \partial \sigma }}&-{\tfrac {{{\partial }^{2}}\Lambda }{\partial {{\sigma }^{2}}}}\\\end{matrix}}\right)_{\mu ={\widehat {\mu }},\sigma ={\widehat {\sigma }}}^{-1}\,\!}$

${\displaystyle \Lambda \,\!}$ is the log-likelihood function of the normal distribution, described in Parameter Estimation and Appendix D.

### Bounds on Reliability

The reliability of the logistic distribution is:

${\displaystyle {\widehat {R}}={\frac {1}{1+{{e}^{\widehat {z}}}}}\,\!}$

where:

${\displaystyle {\widehat {z}}={\frac {t-{\widehat {\mu }}}{\widehat {\sigma }}}\,\!}$

Here ${\displaystyle -\infty , ${\displaystyle -\infty <\mu <\infty \,\!}$, ${\displaystyle 0<\sigma <\infty \,\!}$. Therefore, ${\displaystyle z\,\!}$ also is changing from ${\displaystyle -\infty \,\!}$ to ${\displaystyle +\infty \,\!}$. Then the bounds on ${\displaystyle z\,\!}$ are estimated from:

${\displaystyle {{z}_{U}}={\widehat {z}}+{{K}_{\alpha }}{\sqrt {Var({\widehat {z}}){\text{ }}}}\,\!}$

${\displaystyle {{z}_{L}}={\widehat {z}}-{{K}_{\alpha }}{\sqrt {Var({\widehat {z}}){\text{ }}}}{\text{ }}\,\!}$

where:

${\displaystyle Var({\widehat {z}})={{({\frac {\partial z}{\partial \mu }})}^{2}}Var({\widehat {\mu }})+2({\frac {\partial z}{\partial \mu }})({\frac {\partial z}{\partial \sigma }})Cov({\widehat {\mu }},{\widehat {\sigma }})+{{({\frac {\partial z}{\partial \sigma }})}^{2}}Var({\widehat {\sigma }})\,\!}$

or:

${\displaystyle Var({\widehat {z}})={\frac {1}{{\sigma }^{2}}}(Var({\widehat {\mu }})+2{\widehat {z}}Cov({\widehat {\mu }},{\widehat {\sigma }})+{{\widehat {z}}^{2}}Var({\widehat {\sigma }}))\,\!}$

The upper and lower bounds on reliability are:

${\displaystyle {{R}_{U}}={\frac {1}{1+{{e}^{{z}_{L}}}}}{\text{(upper bound)}}\,\!}$
${\displaystyle {{R}_{L}}={\frac {1}{1+{{e}^{{z}_{U}}}}}{\text{(lower bound)}}\,\!}$

### Bounds on Time

The bounds around time for a given logistic percentile (unreliability) are estimated by first solving the reliability equation with respect to time as follows:

${\displaystyle {\widehat {T}}({\widehat {\mu }},{\widehat {\sigma }})={\widehat {\mu }}+{\widehat {\sigma }}z\,\!}$

where:

{\displaystyle {\begin{aligned}z=\ln(1-R)-\ln(R)\end{aligned}}\,\!}
${\displaystyle Var({\widehat {T}})={{({\frac {\partial T}{\partial \mu }})}^{2}}Var({\widehat {\mu }})+2({\frac {\partial T}{\partial \mu }})({\frac {\partial T}{\partial \sigma }})Cov({\widehat {\mu }},{\widehat {\sigma }})+{{({\frac {\partial T}{\partial \sigma }})}^{2}}Var({\widehat {\sigma }})\,\!}$

or:

${\displaystyle Var({\widehat {T}})=Var({\widehat {\mu }})+2{\widehat {z}}Cov({\widehat {\mu }},{\widehat {\sigma }})+{{\widehat {z}}^{2}}Var({\widehat {\sigma }})\,\!}$

The upper and lower bounds are then found by:

${\displaystyle {{T}_{U}}={\widehat {T}}+{{K}_{\alpha }}{\sqrt {Var({\widehat {T}}){\text{ }}}}({\text{upper bound}})\,\!}$
${\displaystyle {{T}_{L}}={\widehat {T}}-{{K}_{\alpha }}{\sqrt {Var({\widehat {T}}){\text{ }}}}({\text{lower bound}})\,\!}$