methods-lra: Functionality for log-ratio analysis ('lra') objects

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

Functionality for log-ratio analysis ('lra') objects

Description

These methods extract data from, and attribute new data to, objects of class "lra", a class introduced in this package to organize the singular value decomposition of a double-centered log-transformed data matrix output by lra().

Usage

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

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

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

## S3 method for class 'lra'
recover_inertia(x)

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

## S3 method for class 'lra'
recover_conference(x)

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

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

## S3 method for class 'lra'
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 singular value decomposition-based techniques: methods-cancor, methods-correspondence, methods-lda, methods-mca, methods-prcomp, methods-princomp, methods-svd

Examples

# data frame of violent crime arrests in the United States
class(USArrests)
head(USArrests)
# get state abbreviation data
state <- data.frame(
  name = state.name,
  abb = state.abb
)

# compute (non-compositional, unweighted) log-ratio analysis
USArrests %>%
  subset(select = -UrbanPop) %>%
  lra() %>%
  as_tbl_ord() %>%
  print() -> arrests_lra

# augment log-ratio profiles with names and join state abbreviations
arrests_lra %>%
  augment_ord() %>%
  left_join_rows(state, by = "name") %>%
  print() -> arrests_lra

# recover state and arrest profiles
head(get_rows(arrests_lra))
get_cols(arrests_lra)
# initially, inertia is conferred on neither factor
get_conference(arrests_lra)

# row-principal biplot
arrests_lra %>%
  confer_inertia("rows") %>%
  ggbiplot(aes(color = .matrix), sec.axes = "cols", scale.factor = 1/20) +
  scale_color_manual(values = c("tomato4", "turquoise4")) +
  theme_bw() +
  geom_rows_text(aes(label = abb), size = 3, alpha = .75) +
  geom_cols_polygon(fill = NA, linetype = "dashed") +
  geom_cols_text(aes(label = name, size = weight), fontface = "bold") +
  scale_size_area(guide = "none") +
  ggtitle(
    "Non-compositional LRA of violent crime arrest rates",
    "United States, 1973"
  ) +
  expand_limits(x = c(-.35)) +
  guides(color = "none")

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