regression_lin: regression_lin

In janbrederecke/analyzeD: Functions that help to easily analyze data

regression_lin

Description

Calculates linear regression models using lm(). Designed to make analysis with multiple predictors / outcomes easy and quick.

Usage

regression_lin(
  .data,
  .outcomes = NULL,
  .predictors = NULL,
  .covariates = NULL,
  .annotation = NULL,
  .subset = NULL,
  .cpus = 1,
  .sort_by = "outcomes",
  .std_prd = FALSE,
  .std_cov = NULL,
  .summary = FALSE,
  .interaction = NULL,
  .imputed_predictors = FALSE,
  .imputed_outcomes = FALSE,
  ...
)

Arguments

.data	A data.frame or .mids object.
.outcomes	A character vector containing the outcomes.
.predictors	A character vector containing the predictors.
.covariates	A character vector containing covariates.
.annotation	A matrix or data.frame of format (name, pname, unit, short_pname, comment) that contains pretty names for the used variables.
.subset	Can be used to internally subset the data. Use .subset = "variable == 'x'" to subset data.
.cpus	Input number of desired cpus to use. Useful only in case of big datasets and multiple outcomes/predictors.
.sort_by	A character string that indicates either to sort the analyses by "outcomes" or by "predictors".
.std_prd	If TRUE, predictors are standardized.
.std_cov	Character vector of covariates that should be standardized.
.summary	If TRUE, an additional summary of all analyses is returned.
.interaction	Can be used to specify interactions using a list of character vectors containing the interaction variables, e.g. list(c("variable1", "variable2"), c("variable2", "variable3")).
.imputed_predictors	If TRUE, cases with imputed predictors are used.
.imputed_outcomes	If TRUE, cases with imputed outcomes are used.
...	Optional input passed directly to the regression function.

Value

A list containing individual results tables for each regression analysis.

Examples

-

janbrederecke/analyzeD documentation built on Jan. 2, 2023, 5:41 a.m.