adjust {datawizard}

Adjust data for the effect of other variable(s)

Description

This function can be used to adjust the data for the effect of other variables present in the dataset. It is based on an underlying fitting of regressions models, allowing for quite some flexibility, such as including factors as random effects in mixed models (multilevel partialization), continuous variables as smooth terms in general additive models (non-linear partialization) and/or fitting these models under a Bayesian framework. The values returned by this function are the residuals of the regression models. Note that a regular correlation between two "adjusted" variables is equivalent to the partial correlation between them.

Usage

adjust(
  data,
  effect = NULL,
  select = is.numeric,
  exclude = NULL,
  multilevel = FALSE,
  additive = FALSE,
  bayesian = FALSE,
  keep_intercept = FALSE,
  ignore_case = FALSE,
  regex = FALSE,
  verbose = FALSE
)

data_adjust(
  data,
  effect = NULL,
  select = is.numeric,
  exclude = NULL,
  multilevel = FALSE,
  additive = FALSE,
  bayesian = FALSE,
  keep_intercept = FALSE,
  ignore_case = FALSE,
  regex = FALSE,
  verbose = FALSE
)

Arguments

data

A data frame.

effect

Character vector of column names to be adjusted for (regressed out). If NULL (the default), all variables will be selected.

select

Variables that will be included when performing the required tasks. Can be either

• a variable specified as a literal variable name (e.g., column_name),

• a string with the variable name (e.g., "column_name"), or a character vector of variable names (e.g., c("col1", "col2", "col3")),

• a formula with variable names (e.g., ~column_1 + column_2),

• a vector of positive integers, giving the positions counting from the left (e.g. 1 or c(1, 3, 5)),

• a vector of negative integers, giving the positions counting from the right (e.g., -1 or -1:-3),

• one of the following select-helpers: starts_with(), ends_with(), contains(), a range using : or regex(""). starts_with(), ends_with(), and contains() accept several patterns, e.g starts_with("Sep", "Petal").

• or a function testing for logical conditions, e.g. is.numeric() (or is.numeric), or any user-defined function that selects the variables for which the function returns TRUE (like: foo <- function(x) mean(x) > 3),

• ranges specified via literal variable names, select-helpers (except regex()) and (user-defined) functions can be negated, i.e. return non-matching elements, when prefixed with a -, e.g. -ends_with(""), -is.numeric or -(Sepal.Width:Petal.Length). Note: Negation means that matches are excluded, and thus, the exclude argument can be used alternatively. For instance, select=-ends_with("Length") (with -) is equivalent to exclude=ends_with("Length") (no -). In case negation should not work as expected, use the exclude argument instead.

If NULL, selects all columns. Patterns that found no matches are silently ignored, e.g. extract_column_names(iris, select = c("Species", "Test")) will just return "Species".

exclude

See select, however, column names matched by the pattern from exclude will be excluded instead of selected. If NULL (the default), excludes no columns.

multilevel

If TRUE, the factors are included as random factors. Else, if FALSE (default), they are included as fixed effects in the simple regression model.

additive

If TRUE, continuous variables as included as smooth terms in additive models. The goal is to regress-out potential non-linear effects.

bayesian

If TRUE, the models are fitted under the Bayesian framework using rstanarm.

keep_intercept

If FALSE (default), the intercept of the model is re-added. This avoids the centering around 0 that happens by default when regressing out another variable (see the examples below for a visual representation of this).

ignore_case

Logical, if TRUE and when one of the select-helpers or a regular expression is used in select, ignores lower/upper case in the search pattern when matching against variable names.

regex

Logical, if TRUE, the search pattern from select will be treated as regular expression. When regex = TRUE, select must be a character string (or a variable containing a character string) and is not allowed to be one of the supported select-helpers or a character vector of length > 1. regex = TRUE is comparable to using one of the two select-helpers, select = contains("") or select = regex(""), however, since the select-helpers may not work when called from inside other functions (see 'Details'), this argument may be used as workaround.

verbose

Toggle warnings.

Value

A data frame comparable to data, with adjusted variables.

Examples

adjusted_all <- adjust(attitude)
head(adjusted_all)
adjusted_one <- adjust(attitude, effect = "complaints", select = "rating")
head(adjusted_one)

adjust(attitude, effect = "complaints", select = "rating", bayesian = TRUE)
adjust(attitude, effect = "complaints", select = "rating", additive = TRUE)
attitude$complaints_LMH <- cut(attitude$complaints, 3)
adjust(attitude, effect = "complaints_LMH", select = "rating", multilevel = TRUE)


# Generate data
data <- simulate_correlation(n = 100, r = 0.7)
data$V2 <- (5 * data$V2) + 20 # Add intercept

# Adjust
adjusted <- adjust(data, effect = "V1", select = "V2")
adjusted_icpt <- adjust(data, effect = "V1", select = "V2", keep_intercept = TRUE)

# Visualize
plot(
  data$V1, data$V2,
  pch = 19, col = "blue",
  ylim = c(min(adjusted$V2), max(data$V2)),
  main = "Original (blue), adjusted (green), and adjusted - intercept kept (red) data"
)
abline(lm(V2 ~ V1, data = data), col = "blue")
points(adjusted$V1, adjusted$V2, pch = 19, col = "green")
abline(lm(V2 ~ V1, data = adjusted), col = "green")
points(adjusted_icpt$V1, adjusted_icpt$V2, pch = 19, col = "red")
abline(lm(V2 ~ V1, data = adjusted_icpt), col = "red")