These functions are the `summary`

and `print`

methods for objects of type
`glmmNPML`

and `glmmGQ`

.

1 2 3 4 5 6 7 8 | ```
## S3 method for class 'glmmNPML'
summary(object, digits = max(3, getOption("digits") - 3), ...)
## S3 method for class 'glmmGQ'
summary(object, digits = max(3, getOption("digits") - 3), ...)
## S3 method for class 'glmmNPML'
print(x, digits=max(3,getOption('digits')-3), ...)
## S3 method for class 'glmmGQ'
print(x, digits=max(3,getOption('digits')-3), ...)
``` |

`object` |
a fitted object of class |

`x` |
a fitted object of class |

`digits` |
number of digits; applied on various displayed quantities. |

`...` |
further arguments, which will mostly be ignored. |

The `summary...`

- and `print...`

-functions invoke the generic
`UseMethod(...)`

function and detect the right model class
automatically. In other words, it is enough to write
`summary(...)`

or `print(...)`

.

Prints regression output or summary on screen.

Objects returned by `summary.glmmNPML`

or `summary.glmmGQ`

are
essentially identical to objects of class `glmmNPML`

or `glmmGQ`

.
However, their `$coef`

component contains the parameter standard errors
and t values (taken from the GLM fitted in the last EM iteration), and they
have two additional components `$dispersion`

and `$lastglmsumm`

providing the estimated dispersion parameter and a summary of the `glm`

fitted in the last EM iteration.

Please note that the provided parameter standard errors tend to be underestimated as the uncertainty due to the EM algorithm is not incorporated into them. According to Aitkin et al (2009), Section 7.5, page 440, more accurate standard errors can be obtained by dividing the (absolute value of the) parameter estimate through the square root of the change in disparity when omitting/not omitting the variable from the model.

originally from Ross Darnell (2002), modified and prepared for publication by Jochen Einbeck and John Hinde (2007).

Aitkin, M., Francis, B. and Hinde, J. (2009). Statistical Modelling in R. Oxford Statistical Science Series, Oxford, UK.

`alldist`

, `allvc`

, `summary`

,
`print`

, `family.glmmNPML`

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

All documentation is copyright its authors; we didn't write any of that.