This is documentation for Mathematica 8, which was
based on an earlier version of the Wolfram Language.

# CovarianceEstimatorFunction

 CovarianceEstimatorFunction is an option for generalized linear model fitting functions that specifies the estimator for the parameter covariance matrix.
• Possible settings include and which use the expected information matrix and observed information matrix, respectively.
• The covariance matrix is equivalent to , where is the dispersion parameter and is Fisher's information matrix.
Fit a generalized linear model:
Compute the covariance matrix using the expected information matrix:
Use the observed information matrix instead:
Fit a generalized linear model:
 Out[2]=
Compute the covariance matrix using the expected information matrix:
 Out[3]=
Use the observed information matrix instead:
 Out[4]=
 Out[5]=
 Scope   (2)
Specify the covariance estimate within the FittedModel:
Use with LogitModelFit:
Use with ProbitModelFit:
Error estimates and confidence intervals involve covariance estimates:
Estimate errors and intervals using expected information:
Use observed information:
CovarianceEstimatorFunction controls the general structure of the covariance:
DispersionEstimatorFunction affects the scale:
The ratio of the errors squared is the ratio of the dispersion estimates:
New in 7