Support quantile linear regression

[brshallo]

Add support for quantile linear regression, e.g.

library(parsnip)

linear_reg() %>% 
  set_engine("quantreg", tau = 0.50) %>% 
  set_mode("quantiles") # maybe unnecessary... see #85

(This issue stems from Rstudio Community thread and Mara's encouragement to open an issue to move discussion to github.)

[juliasilge]

Related to #119

[SaintRod]

Hi, @juliasilge. I wanted to chime in and add support for adding quantreg as an engine for linear_reg().

Two follow-up questions:

• Is there an existing way to model the median vs the mean?
• If ^ == no, do you think that adding quantreg as an engine is something that could be contributed to parsnip via a PR? I'd be thrilled to try if so...

PS. Thanks for all the great work on tidymodels!

[pbhogale]

I wanted to add my support for this too. Enabling quantile regression with quantreg::rq as well as xgboost would be awesome.

[SaintRod]

I've been pretty eagerly waiting for this and based on the latest tidymodels quarterly blog post... it looks like this feature now exists!

source: https://www.tidyverse.org/blog/2025/02/tidymodels-2025-q1/

An example provided from the above link

ibrary(tidymodels)
tidymodels_prefer()

ames <- 
  modeldata::ames |> 
  mutate(Sale_Price = log10(Sale_Price)) |> 
  select(Sale_Price, Latitude)

quant_spec <- 
  linear_reg() |> 
  set_engine("quantreg") |> 
  set_mode("quantile regression", quantile_levels = c(0.1, 0.5, 0.9))

🥳🥳🥳🥳🥳🥳