stat.anova
Add Statistics Columns to an Anova Table
Description
Creates a new anova table with appropriate statistics columns.
Usage
stat.anova(table, test = c("Chisq", "F", "Cp"), scale, df.scale, n, familyName = "gaussian")
Arguments
| table |
an anova object
|
| test |
what test to use; the choices are "Chisq", "F", or "Cp"
|
| scale |
an estimate of scale, needed for all test.
|
| df.scale |
the (residual) degrees of freedom associated with the scale estimate (as in the denominator of an F-ratio).'
|
| n |
the sample size for the data underlying the anova table.
|
| familyName |
The name of the glm family being used by the object(s) in the anova.
If it is "binomial" or "poisson" and an F-test is requested
then stat.anova will warn that the test is inappropriate (since
there is not an independent estimate of dispersion) and the p-value will
be computed using df2=Inf.
|
Value
a new anova table is returned, with the appropriate statisticscolumns included.
This function relies on the names of the columns being in a small subset of allowed names.
It is mainly intended as a support function for anova(), step(), add1()
or drop1().
References
Hastie, T. J. and Pregibon, D. 1992. Generalized linear models. Statistical Models in S.
Chambers, J. M. and Hastie, T. J. (Eds.) Pacific Grove, CA.: Wadsworth & Brooks/Cole. Chapter 6.
See Also
Examples
numdead <- c(1, 4, 9, 13, 18, 20, 0, 2, 6, 10, 12, 16)
sex <- factor(rep(c("M", "F"), c(6, 6)))
ldose <- rep(0:5, 2)
SF <- cbind(numdead, numalive = 20 - numdead)
budworm.lg.anv <- anova(glm(SF ~ sex*ldose, family = binomial))
stat.anova(budworm.lg.anv, test = "F", scale = 0.8, df.scale = 1.2, n = 12)
stat.anova(budworm.lg.anv, test = "Cp", scale = 0.8, df.scale = 1.2, n = 12)
stat.anova(budworm.lg.anv, test = "Chisq", scale = 0.8, df.scale = 1.2, n = 12)
stat.anova(budworm.lg.anv, test = "Chisq", scale = 1.0, df.scale = 1.2, n = 12)