This function summarizes the result of a fitted serp object in a dataframe.
# S3 method for serp
summary(object, ...)An object of class serp.
Not used. Additional summary arguments.
the matrix of coefficients, standard errors, z-values and p-values.
the deviance for the intercept only model.
the log-likelihood for the intercept only model.
list of penalization information obtained with
slope set to "penalize".
the exponentiated coefficients.
an object of class summary.serp. A list (depending on the type of
slope used) of all model components defined in the serp,
function with additional components listed below.
library(serp)
m <- serp(rating ~ temp + contact, slope = "penalize",
reverse = TRUE, link = "logit", tuneMethod = "user",
lambda = 0, data = wine)
summary(m)
#>
#> call:
#> serp(formula = rating ~ temp + contact, link = "logit", slope = "penalize",
#> tuneMethod = "user", reverse = TRUE, data = wine, lambda = 0)
#>
#> Coefficients:
#> Estimate Std Error z value Pr(>|z|)
#> (Intercept):1 1.2258 0.5570 2.201 0.027757 *
#> (Intercept):2 -1.0330 0.4807 -2.149 0.031638 *
#> (Intercept):3 -3.9464 0.9015 -4.377 1.2e-05 ***
#> (Intercept):4 -19.1843 1079.7764 -0.018 0.985825
#> tempwarm:1 19.2427 3597.5553 0.005 0.995732
#> tempwarm:2 2.1108 0.6007 3.514 0.000441 ***
#> tempwarm:3 2.9404 0.8283 3.550 0.000385 ***
#> tempwarm:4 17.0636 1079.7763 0.016 0.987392
#> contactyes:1 1.6594 1.1786 1.408 0.159167
#> contactyes:2 1.3429 0.5830 2.303 0.021258 *
#> contactyes:3 1.6928 0.6602 2.564 0.010347 *
#> contactyes:4 1.1622 0.9054 1.284 0.199293
#> ---
#> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#>
#> Number of iterations: 18
#>
#> Loglik: -84.61093 on 276 degrees of freedom
#>
#> AIC: 193.2219
#>
#> Exponentiated coefficients:
#> tempwarm:1 tempwarm:2 tempwarm:3 tempwarm:4 contactyes:1 contactyes:2
#> 2.275154e+08 8.255243e+00 1.892362e+01 2.574225e+07 5.256007e+00 3.830073e+00
#> contactyes:3 contactyes:4
#> 5.434719e+00 3.196963e+00
#>
#> Regularization Info:
#> penalty: SERP
#> tuneMethod: user
#> value:
#> lambda: 0