vinereg

R-CMD-check Coverage status CRAN status

Fit D-vine copula regression models for conditional mean and quantile prediction with continuous or discrete variables.

How to install

Functionality

vinereg provides:

See the package website for the complete reference and worked examples.

Example

set.seed(5)
library(vinereg)
data(mtcars)

# declare factors and discrete variables
for (var in c("cyl", "vs", "gear", "carb"))
    mtcars[[var]] <- as.ordered(mtcars[[var]])
mtcars[["am"]] <- as.factor(mtcars[["am"]])

# fit model
(fit <- vinereg(mpg ~ ., family_set = "nonpar", data = mtcars))
#> D-vine regression model: mpg | wt, qsec, drat
#> nobs = 32, edf = 20.35, cll = -57.42, caic = 155.54, cbic = 185.36

summary(fit)
#>    var      edf         cll       caic       cbic      p_value
#> 1  mpg 3.803013 -100.046939 207.699904 213.274116           NA
#> 2   wt 9.871177   29.583463 -39.424574 -24.956036 4.600863e-09
#> 3 qsec 5.389674    7.422915  -4.066482   3.833357 1.449560e-02
#> 4 drat 1.282135    5.617764  -8.671258  -6.791987 1.321129e-03
AIC(fit)
#> [1] 155.5376

# show marginal effects for all selected variables
plot_effects(fit)


# predict mean and median
head(predict(fit, mtcars, alpha = c(NA, 0.5)), 4)
#>       mean      0.5
#> 1 22.57023 22.28455
#> 2 21.76349 21.46394
#> 3 25.51574 25.26815
#> 4 20.18286 20.17140

Vignettes

For more examples, have a look at the vignettes with

vignette("abalone-example", package = "vinereg")
vignette("bike-rental", package = "vinereg")

References

Kraus and Czado (2017). D-vine copula based quantile regression. Computational Statistics & Data Analysis, 110, 1-18. link, preprint

Schallhorn, N., Kraus, D., Nagler, T., Czado, C. (2017). D-vine quantile regression with discrete variables. arXiv preprint, preprint.