Description Usage Arguments Details Value Author(s) References See Also Examples
car package replacements for the summary
, vcov
and confint
functions for lm
objects, with additional arguments and different defaults.
Default methods that call the standard R functions are provided for the Summarize
, Vcov
, and Confint
generics, so the car functions should be safe to use in general.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 | Summarize(object, ...)
## S3 method for class 'lm'
Summarize(object, correlation = FALSE, symbolic.cor = FALSE,
vcov. = Vcov, conf.intervals = FALSE, ...)
## S3 method for class 'Summarize.lm'
print(x, digits = max(3, getOption("digits") - 3),
symbolic.cor = x$symbolic.cor,
signif.stars = getOption("show.signif.stars"),
header=TRUE, resid.summary=FALSE, adj.r.squared=FALSE, brief=FALSE, ...)
## Default S3 method:
Summarize(object, ...)
Vcov(object, ...)
## S3 method for class 'lm'
Vcov(object, ...)
## Default S3 method:
Vcov(object, ...)
Confint(object, ...)
## S3 method for class 'lm'
Confint(object, parm, level = 0.95, vcov.=Vcov, ...)
## Default S3 method:
Confint(object, ...)
|
object |
an object of class |
correlation, symbolic.cor |
see |
x,digits,signif.stars |
see |
parm,level |
see |
vcov. |
either a function for estimating the covariance matrix of the regression coefficients, e.g., |
conf.intervals |
if |
header |
if |
resid.summary |
if |
adj.r.squared |
if |
brief |
if |
... |
additional arguments, for consistency with other summary methods |
All these functions mimic functions in the stats package for summarizing aspects of a linear model.
The Summarize
and Confint
functions add support for the vcov.
argument, which allows specifying a covariance matrix for the regression coefficients other than the usual covariance matrix returned by the function vcov
. This argument may be either the name of a function, so that the call to vcov.(object)
returns a covariance matrix, or else vcov.
is set equal to a covariance matrix. Setting vcov.=hccm
uses 'proposal 3' described by Long and Ervin (2000) for a sandwich coefficient-variance estimator that may be robust against nonconstant variance. Setting vcov. = hccm(object, type = "hc2")
would use the matrix returned by the hccm
function using proposal 2. If vcov. = Boot
, with a required capital 'B
', then the covariance matrix is computed using a case-resampling bootstrap by the call cov(Boot(object)$t)
; see Boot
and hccm
.
When using vcov. = Boot
, we recommend setting conf.intervals = TRUE
, so confidence intervals based on the Bca method will be based on the same bootstrap as used in the summary. Alternatively, users may prefer to save the bootstrap, and this can be accomplished by first setting b1 <- Boot(object)
, and then calling Summarize(object, vcov. = cov(b1$t))
.
The overall F-test in the Summarize
output uses the supplied covariance matrix in a call to the linearHypothesis
function.
The print method has additional arguments to customize the standard summary
output. Standard output is obtained by setting residual.summary=TRUE, adjusted.r.squared=TRUE
with vcov. = vcov
.
The Summarize.lm
function returns a list with all the elements shown at summary.lm
plus two additional items:
vcov. |
The name of the method used to estimate the covariance matrix |
conf.intervals |
if these were requested, the confidence intervals for coefficient estimates |
The Vcov.lm
function returns the usual covariance matrix returned by vcov
.
The Confint.lm
function returns either (1) the output from confint.lm
if vcov. = vcov
; (2) the output from confint(Boot(object))
if vcov. = Boot
; or (3) Wald-type confidence intervals using the supplied covariance matrix for any other choice of vcov.
.
Sanford Weisberg sandy@umn.edu
Long, J. S. and Ervin, L. H. (2000) Using heteroscedasity consistent standard errors in the linear regression model. The American Statistician 54, 217–224.
White, H. (1980) A heteroskedastic consistent covariance matrix estimator and a direct test of heteroskedasticity. Econometrica 48, 817–838.
summary.lm
, confint
, vcov.lm
, hccm
,
Boot
, linearHypothesis
1 2 3 4 5 6 7 8 9 10 11 12 | mod.prestige <- lm(prestige ~ education + income + type, Prestige)
Summarize(mod.prestige, vcov.=hccm)
print(Summarize(mod.prestige, conf.intervals=TRUE), brief=TRUE)
## Not run:
Summarize(mod.prestige, vcov.=Boot)
## End(Not run)
Confint(mod.prestige, vcov.=hccm)
## Not run:
Confint(mod.prestige, vcov.=Boot)
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.