Generate a long table with complete results of linear regression models

Given a dataframe and a set of dependent and independent variables from that dataframe, it generates a linear regression model for each single dependent variable and creates a new table with all slopes and p-values.

Usage

gentab_lm_long(
  df,
  dep,
  form_ind,
  mdl = "lm",
  left_cens = NULL,
  right_cens = NULL,
  var_perc = FALSE,
  base = exp(1),
  FDR = FALSE,
  filter_sign = FALSE,
  pcutoff = 0.05,
  cutPval = FALSE
)

Arguments

df: a dataframe.
dep: a character vector. Each element must correspond to a column name of the df, each of which must contain numeric values.
form_ind: character vector of length 1. Write here the part of the formual of the linear model the ~ (for example: for the linear model dependent ~ independent1 + independent2 + independent3, you should pass here: "independent1 + independent2 + independent3").
mdl: one of the following: "lm", "lmer", or "tobit". For linear model with fixed effects choose "lm" (it uses the function lm); for linear model with mixed effects choose "lmer" (function lmer of the lmerTest package); for linear model with censored variable choose "tobit" (function tobit of the AER package)
left_cens: NULL or a named numeric vector. If mdl is "tobit", indicate here the left-censored values (e.g.: the limit of detection values). In particular, this argument should be set to NULL (no left-censored values) or a named numeric vector whose names must correspond to variables passed in dep.
right_cens: NULL or a named numeric vector. If mdl is "tobit", indicate here the right-censored values (e.g.: the upper limit of detection values). In particular, this argument should be set to NULL (no right-censored values) or a named numeric vector whose names must correspond to variables passed in dep.
var_perc: logical. Besides the beta slopes, do you also want to know the variation percentage? (if so, please, ensure your data are log-transformormed and scaled; consider using the function data_transf of the present package to do this). The variation percentage is calculated as follow: (((base^beta)-1)*100).
base: numerical of length one. If var_perc is TRUE, it is the base of the logarithm used to log-transform the dependent variables.
FDR: logical. If TRUE, after performing the ANOVA, it also correct p-values across the different variables with a false discovery rate multiple comparison correction (method "fdr" of the function p.adjust).
filter_sign: logical. If TRUE, the table will be filtered and only the p-values lower than the value specified in pcutoff will be considered.
pcutoff: a numeric of length 1, must be between 0 and 1. If filter_sign is TRUE, cut-off value of the p-values.
cutPval: logical. If TRUE, it cut the p-values using the cutP function of the present package.

Value

A tibble the results of the t-tests.