Predicted Values

Fit Analyses

After the model has been fit, the predicted values are calculated from the estimated regression equation.

For linear models, the predicted mean vector of the n observation responses is

$\hat{{\mu}} = X{b} = H y$

$\hat{ \mu_{i}} = x_{i}b$

For generalized linear models,

$\hat{ \mu_{i}} = g^{-1} ( {\eta}_{0i} + x_{i}b)$

where ${ {\eta}_{0i}}$ is the offset for the ith observation.

The predicted values are stored in variables named P_yname for each response variable, where yname is the response variable name.