Summarizing Vector Generalized Additive Model Fits

These functions are all methods for class vgam or summary.vgam objects.

summaryvgam(object, dispersion = NULL, digits = options()$digits - 2,
            presid = TRUE, nopredictors = FALSE)
# S3 method for class 'summary.vgam'
show(x, quote = TRUE, prefix = "",
                            digits = options()$digits-2, nopredictors = NULL)

Arguments

object

an object of class "vgam", which is the result of a call to vgam with at least one s term.

x

an object of class "summary.vgam", which is the result of a call to summaryvgam().

dispersion, digits, presid

See summaryvglm.

quote, prefix, nopredictors

See summaryvglm.

Details

This methods function reports a summary more similar to summary.gam() from gam than summary.gam from mgcv. It applies to G1-VGAMs using s and vector backfitting. In particular, an approximate score test for linearity is conducted for each s term—see Section 4.3.4 of Yee (2015) for details. The p-values from this type of test tend to be biased upwards (too large).

Value

summaryvgam returns an object of class "summary.vgam"; see summary.vgam-class.

See also

vgam, summary.glm, summary.lm, summary.gam from mgcv, summarypvgam for P-VGAMs.

Examples

hfit <- vgam(agaaus ~ s(altitude, df = 2), binomialff, data = hunua)
summary(hfit)
#> 
#> Call:
#> vgam(formula = agaaus ~ s(altitude, df = 2), family = binomialff, 
#>     data = hunua)
#> 
#> Name of additive predictor: logitlink(prob) 
#> 
#> (Default) Dispersion Parameter for binomialff family:   1
#> 
#> Residual deviance:  394.9298 on 389.167 degrees of freedom
#> 
#> Log-likelihood: -197.4649 on 389.167 degrees of freedom
#> 
#> Number of Fisher scoring iterations:  6 
#> 
#> DF for Terms and Approximate Chi-squares for Nonparametric Effects
#> 
#>                     Df Npar Df Npar Chisq     P(Chi)
#> (Intercept)          1                              
#> s(altitude, df = 2)  1     0.8     9.2773 0.00167449
summary(hfit)@anova  # Table for (approximate) testing of linearity
#>                     Df Npar Df Npar Chisq      P(Chi)
#> (Intercept)          1      NA         NA          NA
#> s(altitude, df = 2)  1     0.8   9.277346 0.001674495