Confidence Intervals
Confidence intervals are computed for all model parameters and are reported in the
"Analysis of Parameter Estimates" table. The confidence coefficient can be specified
with the ALPHA=
MODEL statement option, resulting in a
two-sided confidence coefficient. The default confidence coefficient is 95%,
corresponding to
.Regression Parameters
A two-sided
confidence interval
for the regression parameter
is based on the asymptotic normality
of the maximum likelihood estimator
and is
computed by


where
is the estimated standard error of
, and
zp is the p×100% percentile of the standard normal distribution.
Scale Parameter
A two-sided
confidence interval
for the scale parameter
in the location-scale model is based on the asymptotic normality
of the logarithm of the maximum-likelihood estimator
, and is computed by
![\sigma_{L}=\hat{\sigma}/\exp[z_{1-\alpha/2}{\rm (SE}_{\hat{\sigma}})/\hat{\sigma}]](images/lfreq15.gif)
![\sigma_{U}=\hat{\sigma}\exp[z_{1-\alpha/2}{\rm (SE}_{\hat{\sigma}})/\hat{\sigma}]](images/lfreq16.gif)
Refer to Meeker and Escobar (1998) for more information.
Weibull Scale and Shape Parameters
The Weibull distribution scale parameter
and shape parameter
are obtained
by transforming the extreme value location parameter
and scale parameter
:


Consequently, two-sided
confidence intervals for the
Weibull scale and shape parameters are computed as
![[\eta_L,\;\; \eta_U ]=[\exp(\mu_L),\;\; \exp(\mu_U)]](images/lfreq22.gif)
![[\beta_L,\;\; \beta_U]=[1/\sigma_U,\;\; 1/\sigma_L]](images/lfreq23.gif)
Gamma Shape Parameter
A two-sided
confidence interval for the 3-parameter
gamma shape parameter
is computed by
![[\delta_L,\;\;\delta_U]=[ \hat{\delta} - z_{1-\alpha/2}({\rm SE}_{\hat{\delta}}), \;\; \hat{\delta} + z_{1-\alpha/2}({\rm SE}_{\hat{\delta}}) ]](images/lfreq25.gif)
Copyright © 2001 by SAS Institute Inc., Cary, NC, USA. All rights reserved.