This function saves rms attributes with the fit object so that
anova.rms, Predict, etc. can be used just as with ols
and other fits. No validate or calibrate methods exist for
Glm though.
Glm(
formula,
family = gaussian,
data = environment(formula),
weights,
subset,
na.action = na.delete,
start = NULL,
offset = NULL,
control = glm.control(...),
model = TRUE,
method = "glm.fit",
x = FALSE,
y = TRUE,
contrasts = NULL,
...
)see stats::glm(); for print x is the result of Glm
ignored
a fit object like that produced by stats::glm() but with
rms attributes and a class of "rms", "Glm",
"glm", and "lm". The g element of the fit object is
the \(g\)-index.
For the print method, format of output is controlled by the user
previously running options(prType="lang") where lang is
"plain" (the default), "latex", or "html".
stats::glm(),Hmisc::GiniMd(), prModFit(), stats::residuals.glm
## Dobson (1990) Page 93: Randomized Controlled Trial :
counts <- c(18,17,15,20,10,20,25,13,12)
outcome <- gl(3,1,9)
treatment <- gl(3,3)
f <- glm(counts ~ outcome + treatment, family=poisson())
f
#>
#> Call: glm(formula = counts ~ outcome + treatment, family = poisson())
#>
#> Coefficients:
#> (Intercept) outcome2 outcome3 treatment2 treatment3
#> 3.045e+00 -4.543e-01 -2.930e-01 6.972e-16 8.237e-16
#>
#> Degrees of Freedom: 8 Total (i.e. Null); 4 Residual
#> Null Deviance: 10.58
#> Residual Deviance: 5.129 AIC: 56.76
anova(f)
#> Analysis of Deviance Table
#>
#> Model: poisson, link: log
#>
#> Response: counts
#>
#> Terms added sequentially (first to last)
#>
#>
#> Df Deviance Resid. Df Resid. Dev Pr(>Chi)
#> NULL 8 10.5814
#> outcome 2 5.4523 6 5.1291 0.06547 .
#> treatment 2 0.0000 4 5.1291 1.00000
#> ---
#> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
summary(f)
#>
#> Call:
#> glm(formula = counts ~ outcome + treatment, family = poisson())
#>
#> Coefficients:
#> Estimate Std. Error z value Pr(>|z|)
#> (Intercept) 3.045e+00 1.709e-01 17.815 <2e-16 ***
#> outcome2 -4.543e-01 2.022e-01 -2.247 0.0246 *
#> outcome3 -2.930e-01 1.927e-01 -1.520 0.1285
#> treatment2 6.972e-16 2.000e-01 0.000 1.0000
#> treatment3 8.237e-16 2.000e-01 0.000 1.0000
#> ---
#> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#>
#> (Dispersion parameter for poisson family taken to be 1)
#>
#> Null deviance: 10.5814 on 8 degrees of freedom
#> Residual deviance: 5.1291 on 4 degrees of freedom
#> AIC: 56.761
#>
#> Number of Fisher Scoring iterations: 4
#>
f <- Glm(counts ~ outcome + treatment, family=poisson())
# could have had rcs( ) etc. if there were continuous predictors
f
#> General Linear Model
#>
#> Glm(formula = counts ~ outcome + treatment, family = poisson())
#>
#> Model Likelihood
#> Ratio Test
#> Obs 9 LR chi2 5.45
#> Residual d.f.4 d.f. 4
#> g 0.227 Pr(> chi2) 0.2440
#>
#> Coef S.E. Wald Z Pr(>|Z|)
#> Intercept 3.0445 0.1709 17.81 <0.0001
#> outcome=2 -0.4543 0.2022 -2.25 0.0246
#> outcome=3 -0.2930 0.1927 -1.52 0.1285
#> treatment=2 0.0000 0.2000 0.00 1.0000
#> treatment=3 0.0000 0.2000 0.00 1.0000
#>
anova(f)
#> Wald Statistics Response: counts
#>
#> Factor Chi-Square d.f. P
#> outcome 5.49 2 0.0643
#> treatment 0.00 2 1.0000
#> TOTAL 5.49 4 0.2409
summary(f, outcome=c('1','2','3'), treatment=c('1','2','3'))
#> Effects Response : counts
#>
#> Factor Low High Diff. Effect S.E. Lower 0.95 Upper 0.95
#> outcome - 1:2 2 1 NA 4.5426e-01 0.20217 -0.10706 1.01560
#> outcome - 3:2 2 3 NA 1.6127e-01 0.21512 -0.43600 0.75854
#> treatment - 1:2 2 1 NA -6.9715e-16 0.20000 -0.55529 0.55529
#> treatment - 3:2 2 3 NA 1.2657e-16 0.20000 -0.55529 0.55529
#>