Template:Appr conf bounds for arr-exp: Difference between revisions

From ReliaWiki
Jump to navigation Jump to search
(Created page with '==Approximate Confidence Bounds for the Arrhenius-Exponential== <br> There are different methods for computing confidence bounds. ALTA utilizes confidence bounds that are based o…')
 
 
(4 intermediate revisions by 2 users not shown)
Line 1: Line 1:
==Approximate Confidence Bounds for the Arrhenius-Exponential==
#REDIRECT [[Arrhenius_Relationship#Approximate_Confidence_Bounds_for_the_Arrhenius-Exponential]]
<br>
There are different methods for computing confidence bounds. ALTA utilizes confidence bounds that are based on the asymptotic theory for maximum likelihood estimates, most commonly referred to as the Fisher matrix bounds.
<br>
===Confidence Bounds on the Mean Life===
<br>
The Arrhenius-exponential distribution is given by Eqn. (arrhenius) by setting  <math>m=L(V)</math>  as shown in Eqn. (Arrean). The upper  <math>({{m}_{U}})</math>  and lower  <math>({{m}_{L}})</math>  bounds on the mean life are then estimated by:
 
<br>
::<math>\begin{align}
  & {{m}_{U}}= & \widehat{m}\cdot {{e}^{\tfrac{{{K}_{\alpha }}\sqrt{Var(\widehat{m})}}{\widehat{m}}}} \\
& {{m}_{L}}= & \widehat{m}\cdot {{e}^{-\tfrac{{{K}_{\alpha }}\sqrt{Var(\widehat{m})}}{\widehat{m}}}} 
\end{align}</math>
 
<br>
where  <math>{{K}_{\alpha }}</math>  is defined by:
 
<br>
::<math>\alpha =\frac{1}{\sqrt{2\pi }}\mathop{}_{{{K}_{\alpha }}}^{\infty }{{e}^{-\tfrac{{{t}^{2}}}{2}}}dt=1-\Phi ({{K}_{\alpha }})</math>
 
<br>
If  <math>\delta </math>  is the confidence level (i.e., 95%=0.95), then  <math>\alpha =\tfrac{1-\delta }{2}</math>  for the two-sided bounds, and  <math>\alpha =1-\delta </math>  for the one-sided bounds. The variance of  <math>\widehat{m}</math>  is given by:
 
<br>
::<math>\begin{align}
  & Var(\widehat{m})= & {{\left( \frac{\partial m}{\partial C} \right)}^{2}}Var(\widehat{C})+{{\left( \frac{\partial m}{\partial B} \right)}^{2}}Var(\widehat{B}) \\
&  & +2\left( \frac{\partial m}{\partial C} \right)\left( \frac{\partial m}{\partial B} \right)Cov(\widehat{B},\widehat{C}) 
\end{align}</math>
 
<br>
or:
 
<br>
::<math>Var(\widehat{m})={{e}^{\tfrac{2\widehat{B}}{V}}}\left[ Var(\widehat{C})+\frac{{{\widehat{C}}^{2}}}{{{V}^{2}}}Var(\widehat{B})+\frac{2\widehat{C}}{V}Cov(\widehat{B},\widehat{C}) \right]</math>
 
<br>
The variances and covariance of  <math>B</math>  and  <math>C</math>  are estimated from the local Fisher matrix (evaluated at  <math>\widehat{B}</math> ,  <math>\widehat{C})</math>  as follows:
 
<br>
::<math>\left[ \begin{matrix}
  Var(\widehat{B}) & Cov(\widehat{B},\widehat{C})  \\
  Cov(\widehat{C},\widehat{B}) & Var(\widehat{C})  \\
\end{matrix} \right]={{\left[ \begin{matrix}
  -\tfrac{{{\partial }^{2}}\Lambda }{\partial {{B}^{2}}} & -\tfrac{{{\partial }^{2}}\Lambda }{\partial B\partial C}  \\
  -\tfrac{{{\partial }^{2}}\Lambda }{\partial C\partial B} & -\tfrac{{{\partial }^{2}}\Lambda }{\partial {{C}^{2}}}  \\
\end{matrix} \right]}^{-1}}</math>
 
===Confidence Bounds on Reliability===
<br>
The bounds on reliability for any given time,  <math>T</math> , are estimated by:
<br>
::<math>\begin{align}
  & {{R}_{U}}(T)= & {{e}^{-\tfrac{T}{{{m}_{U}}}}} \\
& {{R}_{L}}(T)= & {{e}^{-\tfrac{T}{{{m}_{L}}}}} 
\end{align}</math>
 
<br>
where  <math>{{m}_{U}}</math>  and  <math>{{m}_{L}}</math>  are estimated using Eqns. (ArrhuUpper) and (ArrhuLower).
<br>
 
===Confidence Bounds on Time===
<br>
The bounds on time (ML estimate of time) for a given reliability are estimated by first solving the reliability function with respect to time:
 
<br>
::<math>\widehat{T}=-\widehat{m}\cdot \ln (R)</math>
 
<br>
The corresponding confidence bounds are then estimated from:
 
<br>
::<math>\begin{align}
  & {{T}_{U}}= & -{{m}_{U}}\cdot \ln (R) \\
& {{T}_{L}}= & -{{m}_{L}}\cdot \ln (R) 
\end{align}</math>
 
 
where  <math>{{m}_{U}}</math>  and  <math>{{m}_{L}}</math>  are estimated using Eqns. (ArrhuUpper) and (ArrhuLower).

Latest revision as of 05:47, 16 August 2012