xpose.xtras (version 0.2.2)

shk_vs_contcov: Shrinkage contribution versus continuous covariates

Description

Mirrors eta_vs_contcov(), but for the per-individual shrinkage contribution diagnostic (shk type columns, see derive_shk()/backfill_shk()) instead of the etas themselves.

Usage

shk_vs_contcov(
  xpdb,
  mapping = NULL,
  shkvar = NULL,
  covvar = NULL,
  drop_fixed = TRUE,
  linsm = FALSE,
  type = "ps",
  list = TRUE,
  title = "Shrinkage contribution versus continuous covariates | @run",
  subtitle = "Based on @nind individuals",
  caption = "@dir",
  tag = NULL,
  log = NULL,
  guide = TRUE,
  facets,
  .problem,
  quiet,
  ...
)

Value

The desired plot, or (when shkvar resolves to more than one shk column and list = TRUE) a plain list of one plot per column.

Arguments

xpdb

<xp_xtras> or <xpose_data> object

mapping

ggplot2 style mapping

shkvar

tidyselect for shk variables

covvar

tidyselect for continuous covariate variables; NULL (default) selects every continuous covariate in the xpdb data index.

drop_fixed

As in xpose

linsm

If type contains "s" should the smooth method by lm?

type

Passed to xplot_scatter

list

<logical> Only relevant when shkvar resolves to more than one shk column. If TRUE (default, for backwards compatibility), returns a plain list of one plot per shk column. If FALSE, they are instead combined onto one shared plot -- faceted by shk column, in addition to the existing per-covariate facet -- automatically paginating (at most 9 panels per page, i.e. ncol/nrow of 3) via xpose's own facet_wrap_paginate mechanism. Printing the returned plot renders every page; pass page to print() to select a specific one.

title

Plot title

subtitle

Plot subtitle

caption

Plot caption

tag

Plot tag

log

Log scale covariate value?

guide

Add guide line?

facets

Additional facets

.problem

Problem number

quiet

Silence output

...

Any additional aesthetics.

Examples

# \donttest{

xpdb_x %>%
  backfill_shk() %>%
  shk_vs_contcov()

# Combine all shk columns onto one shared, faceted plot instead of a list
xpdb_x %>%
  backfill_shk() %>%
  shk_vs_contcov(list = FALSE)
# }