Summarizing Generalized Linear Model Fits
These functions are all
methods for class
## S3 method for class 'glm': summary(object, dispersion = NULL, correlation = FALSE, symbolic.cor = FALSE, \dots)
## S3 method for class 'summary.glm': print(x, digits = max(3, getOption("digits") - 3), symbolic.cor = x$symbolic.cor, signif.stars = getOption("show.signif.stars"), ...)
- an object of class
"glm", usually, a result of a call to
- an object of class
"summary.glm", usually, a result of a call to
- the dispersion parameter for the family used.
Either a single numerical value or
NULL(the default), when it is inferred from
- 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, print the correlations in a symbolic form (see
symnum) rather than as numbers.
- logical. If
significance starsare printed for each coefficient.
- further arguments passed to or from other methods.
print.summary.glm tries to be smart about formatting the
coefficients, standard errors, etc. and additionally gives
coefficients component of the result gives the estimated
coefficients and their estimated standard errors, together with their
ratio. This third column is labelled
t ratio if the
dispersion is estimated, and
z ratio if the dispersion is known
(or fixed by the family). A fourth column gives the two-tailed
p-value corresponding to the t or z ratio based on a Student t or
Normal reference distribution. (It is possible that the dispersion is
not known and there are no residual degrees of freedom from which to
estimate it. In that case the estimate is
Aliased coefficients are omitted in the returned object but restored
Correlations are printed to two decimal places (or symbolically): to
see the actual correlations print
The dispersion of a GLM is not used in the fitting process, but it is
needed to find standard errors.
dispersion is not supplied or
the dispersion is taken as
1 for the
Poisson families, and otherwise estimated by the residual
Chisquared statistic (calculated from cases with non-zero weights)
divided by the residual degrees of freedom.
summary can be used with Gaussian
glm fits to handle the
case of a linear regression with known error variance, something not
summary.glmreturns an object of class
"summary.glm", a list with components
call the component from
family the component from
deviance the component from
contrasts the component from
df.residual the component from
null.deviance the component from
df.null the component from
deviance.resid the deviance residuals: see
coefficients the matrix of coefficients, standard errors, z-values and p-values. Aliased coefficients are omitted. aliased named logical vector showing if the original coefficients are aliased. dispersion either the supplied argument or the inferred/estimated dispersion if the latter is
df a 3-vector of the rank of the model and the number of residual degrees of freedom, plus number of coefficients (including aliased ones). cov.unscaled the unscaled (
dispersion = 1) estimated covariance matrix of the estimated coefficients.
cov.scaled ditto, scaled by
correlation (only if
correlationis true.) The estimated correlations of the estimated coefficients.
symbolic.cor (only if
correlationis true.) The value of the argument
## For examples see example(glm)