methods-lm: Functionality for linear model objects

In ordr: A 'tidyverse' Extension for Ordinations and Biplots

Functionality for linear model objects

Description

These methods extract data from, and attribute new data to, objects of class "lm", "glm", and "mlm" as returned by stats::lm() and stats::glm().

Usage

## S3 method for class 'lm'
as_tbl_ord(x)

## S3 method for class 'lm'
recover_rows(x)

## S3 method for class 'lm'
recover_cols(x)

## S3 method for class 'lm'
recover_coord(x)

## S3 method for class 'lm'
recover_aug_rows(x)

## S3 method for class 'lm'
recover_aug_cols(x)

## S3 method for class 'lm'
recover_aug_coord(x)

## S3 method for class 'glm'
recover_aug_rows(x)

## S3 method for class 'mlm'
recover_rows(x)

## S3 method for class 'mlm'
recover_cols(x)

## S3 method for class 'mlm'
recover_coord(x)

## S3 method for class 'mlm'
recover_aug_rows(x)

## S3 method for class 'mlm'
recover_aug_cols(x)

## S3 method for class 'mlm'
recover_aug_coord(x)

Arguments

x	An ordination object.

Value

The recovery generics recover_*() return core model components, distribution of inertia, supplementary elements, and intrinsic metadata; but they require methods for each model class to tell them what these components are.

The generic as_tbl_ord() returns its input wrapped in the 'tbl_ord' class. Its methods determine what model classes it is allowed to wrap. It then provides 'tbl_ord' methods with access to the recoverers and hence to the model components.

See Also

Other methods for idiosyncratic techniques: methods-kmeans

Other models from the stats package: methods-cancor, methods-cmds, methods-factanal, methods-kmeans, methods-prcomp, methods-princomp

Examples

# Motor Trend design and performance data
head(mtcars)
# regression analysis of performance measures on design specifications
mtcars_centered <- scale(mtcars, scale = FALSE)
mtcars_centered %>%
  as.data.frame() %>%
  lm(formula = mpg ~ wt + cyl) %>%
  print() -> mtcars_lm

# wrap as a 'tbl_ord' object
(mtcars_lm_ord <- as_tbl_ord(mtcars_lm))
# augment everything with names, predictors with observation stats
augment_ord(mtcars_lm_ord)
# calculate influences as the squares of weighted residuals
mutate_rows(augment_ord(mtcars_lm_ord), influence = wt.res^2)

# regression biplot with performance isolines
mtcars_lm_ord %>%
  augment_ord() %>%
  mutate_cols(center = attr(mtcars_centered, "scaled:center")[name]) %>%
  mutate_rows(influence = wt.res^2) %T>% print() %>%
  ggbiplot(aes(x = wt, y = cyl, intercept = `(Intercept)`)) +
  #theme_biplot() +
  geom_origin(marker = "circle", radius = unit(0.02, "snpc")) +
  geom_rows_point(aes(color = influence)) +
  geom_cols_vector() +
  geom_cols_isoline(aes(center = center), by = .5, hjust = -.1) +
  ggtitle(
    "Weight isolines with data colored by importance",
    "Regressing gas mileage onto weight and number of cylinders"
  )

ordr documentation built on Oct. 21, 2022, 1:07 a.m.