An R package for D-vine copula based mean and quantile regression.
-
the stable release from CRAN:
install.packages("vinereg")
-
the latest development version:
# install.packages("devtools") devtools::install_github("tnagler/vinereg", build_vignettes = TRUE)
See the package website.
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 ~ ., data = mtcars))
#> D-vine regression model: mpg | disp, hp, gear, carb, cyl, am.1, wt, vs, qsec, drat
#> nobs = 32, edf = 95, cll = -56.09, caic = 302.19, cbic = 441.43
summary(fit)
#> var edf cll caic cbic p_value
#> 1 mpg 3.2e-11 -98.59271949 197.185439 197.185439 NA
#> 2 disp 2.0e+00 29.53428159 -55.068563 -52.137091 1.490817e-13
#> 3 hp 3.0e+00 2.33231128 1.335377 5.732585 1.980680e-01
#> 4 gear 5.0e+00 2.16379041 5.672419 13.001099 5.032784e-01
#> 5 carb 7.0e+00 2.25663902 9.486722 19.746873 7.191178e-01
#> 6 cyl 9.0e+00 1.57187334 14.856253 28.047876 9.583219e-01
#> 7 am.1 1.0e+01 1.75059329 16.498813 31.156172 9.670581e-01
#> 8 wt 1.3e+01 1.62623809 22.747524 41.802091 9.968597e-01
#> 9 vs 1.3e+01 0.52958111 24.940838 43.995405 9.999946e-01
#> 10 qsec 1.6e+01 0.70430954 30.591381 54.043155 9.999992e-01
#> 11 drat 1.7e+01 0.02886845 33.942263 58.859773 1.000000e+00
# show marginal effects for all selected variables
plot_effects(fit)
#> `geom_smooth()` using method = 'loess' and formula 'y ~ x'
# predict mean and median
head(predict(fit, mtcars, alpha = c(NA, 0.5)), 4)
#> mean 0.5
#> 1 19.34836 19.36129
#> 2 19.17641 19.19810
#> 3 25.28064 25.13942
#> 4 19.70841 19.67779
For more examples, have a look at the vignettes with
vignette("abalone-example", package = "vinereg")
vignette("bike-rental", package = "vinereg")
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. Working paper, preprint.