These functions are all methods
for class addreg
or summary.addreg
objects.
# S3 method for addreg
summary(object, correlation = FALSE, ...)# S3 method for summary.addreg
print(x, digits = max(3L, getOption("digits") - 3L),
signif.stars = getOption("show.signif.stars"), ...)
an object of class "addreg"
, usually from a call to addreg
or addreg.smooth
.
an object of class "summary.addreg"
, usually from a call to
summary.addreg
.
logical; if TRUE
, the correlation matrix of the estimated parameters is
returned and printed.
the number of significant digits to use when printing.
logical; if TRUE
, `significance stars' are printed for each coefficient.
further arguments passed to or from other methods.
summary.addreg
returns an object of class "summary.addreg"
, a list with components
the component from object
.
the component from object
.
the component from object
.
the component from object
.
the component from object
.
the component from object
.
the component from object
.
the component from object
.
the component from object
.
the deviance residuals: see residuals.glm
.
the matrix of coefficients, standard errors, z-values and p-values.
included for compatibility --- always FALSE
.
the inferred/estimated dispersion.
included for compatibility --- a 3-vector of the number of coefficients, the number of residual degrees of freedom, and the number of coefficients (again).
the unscaled (dispersion = 1
) estimated covariance
matrix of the estimated coefficients. NaN
if object$boundary == TRUE
.
ditto, scaled by dispersion
.
if correlation
is TRUE
, the estimated correlations
of the estimated coefficients. NaN
if object$boundary == TRUE
.
For negative binomial models, the object also contains
the estimate of \(\phi\) (scale
-1).
the estimated variance of phi
.
These perform the same function as summary.glm
and print.summary.glm
,
producing similar results for addreg
models. print.summary.addreg
additionally prints
the small-sample corrected AIC (aic.c
), the number of EM iterations for the parameterisation
corresponding to the MLE, and for negative binomial models, the estimate of \(\phi\) (scale
-1)
and its standard error.
The dispersion used in calculating standard errors is fixed as \(1\) for binomial and Poisson models, and is estimated via maximum likelihood for negative binomial models.
# NOT RUN {
## For an example, see example(addreg)
# }
Run the code above in your browser using DataLab