Type:	Package
Title:	Bivariate Automatic Analysis
Version:	1.0.1
Date:	2025-03-03
Maintainer:	José Andrés Flores-García <andres.flores@uaslp.mx>
Description:	Simplify bivariate and regression analyses by automating result generation, including summary tables, statistical tests, and customizable graphs. It supports tests for continuous and dichotomous data, as well as stepwise regression for linear, logistic, and Firth penalized logistic models. While not a substitute for tailored analysis, 'BiVariAn' accelerates workflows and is expanding features like multilingual interpretations of results.The methods for selecting significant statistical tests, as well as the predictor selection in prediction functions, can be referenced in the works of Marc Kery (2003) <doi:10.1890/0012-9623(2003)84[92:NORDIG]2.0.CO;2> and Rainer Puhr (2017) <doi:10.1002/sim.7273>.
License:	GPL (≥ 3)
URL:	https://github.com/AndresFloresG/BiVariAn
BugReports:	https://github.com/AndresFloresG/BiVariAn/issues
Imports:	car, DescTools, dplyr, epitools, fastDummies, ggplot2, ggprism, glue, lifecycle, logistf, magrittr, rlang, rrtable, scales, stats, table1, tidyr, utils
Suggests:	knitr, riskCommunicator, rmarkdown, spelling, testthat (≥ 3.0.0)
VignetteBuilder:	knitr
Config/testthat/edition:	3
Encoding:	UTF-8
Language:	en-US
RoxygenNote:	7.3.2
NeedsCompilation:	no
Packaged:	2025-03-03 15:07:36 UTC; andre
Author:	José Andrés Flores-García [cre, aut, cph], Antonio Augusto Gordillo-Moscoso [aut], Jhoselin Marian Castro-Rodriguez [aut]
Repository:	CRAN
Date/Publication:	2025-03-05 13:10:02 UTC

BiVariAn: Bivariate Automatic Analysis

Description

Simplify bivariate and regression analyses by automating result generation, including summary tables, statistical tests, and customizable graphs. It supports tests for continuous and dichotomous data, as well as stepwise regression for linear, logistic, and Firth penalized logistic models. While not a substitute for tailored analysis, 'BiVariAn' accelerates workflows and is expanding features like multilingual interpretations of results.The methods for selecting significant statistical tests, as well as the predictor selection in prediction functions, can be referenced in the works of Marc Kery (2003) doi:10.1890/0012-9623(2003)84[92:NORDIG]2.0.CO;2 and Rainer Puhr (2017) doi:10.1002/sim.7273.

Author(s)

Maintainer: José Andrés Flores-García andres.flores@uaslp.mx (ORCID) [copyright holder]

Authors:

• Antonio Augusto Gordillo-Moscoso (ORCID)

• Jhoselin Marian Castro-Rodriguez (ORCID)

See Also

Useful links:

• https://github.com/AndresFloresG/BiVariAn

• Report bugs at https://github.com/AndresFloresG/BiVariAn/issues

Automatic generation of barplot with percentages

Description

Automatically generates barplot stratified by group variables with or without percentages.

Usage

auto_bar_categ(
  data,
  groupvar = NULL,
  bar_args = list(),
  theme_func = theme_serene,
  lang_labs = c("EN", "SPA"),
  showpercent = TRUE
)

Arguments

Value

Returns a list containing all barplots as ggplot object. Can be accessed via $ operator

Examples

data<-data.frame(categ = rep(letters[1:2], 10),
var1 = rep(LETTERS[4:5], 10),
var2 = rep(LETTERS[6:7], 10),
var3 = rep(LETTERS[8:9], 10),
var4 = rep(LETTERS[10:11], 10))

data$categ <- as.factor(data$categ)
data$var1 <- as.factor(data$var1)
data$var2 <- as.factor(data$var2)
data$var3 <- as.factor(data$var3)
data$var4 <- as.factor(data$var4)

barplot_list<-auto_bar_categ(data = data, groupvar = "categ", lang_labs = "EN")

barplot_list$var1


# Example using `groupvar` argument as `NULL`
auto_bar_categ(data = data)$var2

Automatic barplot of continous variables

Description

Generates bar plots of contiuous variables based on numerical variables from a data frame. Internally, the function creates a tibble to summarize the data from each variable.

Usage

auto_bar_cont(
  data,
  groupvar,
  err_bar_show = TRUE,
  err_bar = c("sd", "se"),
  col_args = list(),
  lang_labs = c("EN", "SPA"),
  theme_func = theme_serene
)

Arguments

Value

Returns a list containing barplots as ggplot2 objects. Objects can be accessed via $ operator.

Examples

data <- data.frame(group = rep(letters[1:2], 30),
var1 = rnorm(30, mean = 15, sd = 5),
var2 = rnorm(30, mean = 20, sd = 2),
var3 = rnorm(30, mean = 10, sd = 1),
var4 = rnorm(30, mean = 5, sd =2))

data$group<-as.factor(data$group)

# Create a list containing all the plots
barcontplots<-auto_bar_cont(data = data, groupvar = 'group', err_bar = "se", lang_labs = 'EN')

# call to show all storaged plots
barcontplots

# call to show one individual plots
barcontplots$var1

auto_bp_cont

Description

Automatically generates boxplot plots of continuous variables from a database and a grouping variable. The names of the variables are set to the names defined in the database. As a result, graphs generated with the default theme "theme_serene" will be obtained. In this function, the user must define each variable label with "label" function from "table1" package.

Usage

auto_bp_cont(
  data,
  groupvar,
  boxplot_args = list(),
  theme_func = theme_serene,
  lang_labs = c("EN", "SPA")
)

Arguments

Value

A list containing ggplot2 objects with generated plots. Each element can be accessed by using $ operator.

Author(s)

JMCR

Examples

data <- data.frame(group = rep(letters[1:2], 30),
var1 = rnorm(30, mean = 15, sd = 5),
var2 = rnorm(30, mean = 20, sd = 2),
var3 = rnorm(30, mean = 10, sd = 1),
var4 = rnorm(30, mean = 5, sd =2))

data$group<-as.factor(data$group)

# Create a list containing all the plots
boxplots<-auto_bp_cont(data = data, groupvar = 'group', lang_labs = 'EN')

# call to show all storaged plots
boxplots

# call to show one individual plots
boxplots$var1

Generates automatic scatterplot with correlation plot

Description

Automatically generates correlation plots of continuous variables from a database and a reference variable. The names of the variables are set to the names defined in the database. As a result, graphs generated with the default theme "theme_serene" will be obtained. In this function, the user must define each variable label with "label" function from "table1" package

Usage

auto_corr_cont(
  data,
  referencevar = NULL,
  point_args = list(),
  smooth_args = list(),
  theme_func = theme_serene,
  lang_labs = c("EN", "SPA")
)

Arguments

Value

Returns a list containing barplots as ggplot2 objects. Objects can be accessed via $ operator.

Author(s)

JMCR

Examples

data <- data.frame(group = rep(letters[1:2], 30),
var1 = rnorm(30, mean = 15, sd = 5),
var2 = rnorm(30, mean = 20, sd = 2),
var3 = rnorm(30, mean = 10, sd = 1),
var4 = rnorm(30, mean = 5, sd =2))

cont_corrplot <- auto_corr_cont(data = data, referencevar = "var1", lang_labs = "EN")

# Call to show all storaged plots
cont_corrplot

# Call to show one individual plot
cont_corrplot$var2

auto_dens_cont

Description

#' Automatically generates density plots of continuous variables from a database. The names of the variables are set to the names defined in the database. As a result, graphs generated with the default theme "theme_serene" will be obtained. In this function, the user must define each variable label with "label" function from "table1" package.

Usage

auto_dens_cont(
  data,
  s_mean = TRUE,
  s_median = TRUE,
  mean_line_args = list(),
  median_line_args = list(),
  densplot_args = list(),
  theme_func = theme_serene,
  lang_labs = c("EN", "SPA")
)

Arguments

Value

Returns a list containing the generated density plots

Author(s)

JMCR

Examples

data <- data.frame(group = rep(letters[1:2], 30),
var1 = rnorm(30, mean = 15, sd = 5),
var2 = rnorm(30, mean = 20, sd = 2),
var3 = rnorm(30, mean = 10, sd = 1),
var4 = rnorm(30, mean = 5, sd =2))

data$group<-as.factor(data$group)

densityplots <- auto_dens_cont(data = data)

densityplots

densityplots$var1

Automatic generation of pieplots

Description

Generates pie plots based on categorical variables of a data frame.

Usage

auto_pie_categ(
  data,
  pie_bar_args = list(),
  theme_func = theme_serene_void,
  lang_labs = c("EN", "SPA"),
  statistics = TRUE,
  stat_lab = c("percent", "freq"),
  fill_grey = TRUE
)

Arguments

Value

Returns a list containing barplots as ggplot2 objects. Objects can be accessed via $ operator.

Examples

data <- data.frame(categ = rep(c("Categ1", "Categ2"), 25),
var1 = rbinom(50, 2, prob = 0.3),
var2 = rbinom(50, 2, prob = 0.8),
var3 = rbinom(50, 2, prob = 0.7))
data$categ <- as.factor(data$categ)
data$var1 <- as.factor(data$var1)
data$var2 <- as.factor(data$var2)
data$var3 <- as.factor(data$var3)

pieplot_list <- auto_pie_categ(data = data)

# Call for all listed plots
pieplot_list

# Call for one specific plot
pieplot_list$var1

Automatic Shapiro-Wilk test table

Description

Generates a HTML table of raw data from a numerical variables of a dataframe.

Usage

auto_shapiro_raw(data, flextableformat = TRUE)

Arguments

Value

Flextable or dataframe with shapiro wilks results.

Author(s)

JAFG

Examples

auto_shapiro_raw(iris)

auto_viol_cont

Description

Automatically generates violinplots of continuous variables from a database and a grouping variable. The names of the variables are set to the names defined in the database. As a result, graphs generated with the default theme "theme_serene" will be obtained. In this function it is not possible to use labels for the variables, use "auto_viol_cont_wlabels" instead.

Usage

auto_viol_cont(
  data,
  groupvar,
  violinplot_args = list(),
  theme_func = theme_serene,
  lang_labs = c("EN", "SPA")
)

Arguments

Value

Returns a list containing barplots as ggplot2 objects. Objects can be accessed via $ operator.

Author(s)

JMCR

Examples

data <- data.frame(group = rep(letters[1:2], 30),
var1 = rnorm(30, mean = 15, sd = 5),
var2 = rnorm(30, mean = 20, sd = 2),
var3 = rnorm(30, mean = 10, sd = 1),
var4 = rnorm(30, mean = 5, sd =2))

data$group<-as.factor(data$group)

# Create a list containing all the plots
violinplots<-auto_viol_cont(data = data, groupvar = 'group', lang_labs = 'EN')

# call to show all storaged plots
violinplots

# call to show one individual plots
violinplots$var1

Bivariate analysis for 2 groups

Description

Automatic test for continuous variables for 2 groups. Variable names can be assigned using table1::label() function.

Usage

continuous_2g(
  data,
  groupvar,
  ttest_args = list(),
  wilcox_args = list(),
  flextableformat = TRUE
)

Arguments

Value

Returns a dataframe or flextable of 2 groups 2 sided Mann Whitney's U or T test, along with Shapiro-Wilk's p values and Levene's p value.

Examples

 df <- mtcars
 df$am <- as.factor(df$am)
 continuous_2g(data = df,
 groupvar = "am",
 flextableformat = FALSE)

# Set names to variables
if(requireNamespace("table1")){
table1::label(df$mpg) <- "Miles per gallon"
table1::label(df$cyl) <- "Number of cylinders"
table1::label(df$disp) <- "Displacement"
table1::label(df$hp) <- "Gross horsepower"
table1::label(df$drat) <- "Rear axle ratio"

continuous_2g(data = df, groupvar = "am", flextableformat = FALSE)
}

Bivariate analysis for 2 groups for paired data

Description

Automatic paired test for continuous variables for 2 groups. Variable names can be assigned using table1::label() function.

Usage

continuous_2g_pair(
  data,
  groupvar,
  ttest_args = list(),
  wilcox_args = list(),
  flextableformat = TRUE
)

Arguments

Value

A dataframe or flextable with containing p values for paired tests along with statistics for normality and homocedasticity.

Examples

data <- data.frame(group = rep(letters[1:2], 30),
                   var1 = rnorm(60, mean = 15, sd = 5),
                  var2 = rnorm(60, mean = 20, sd = 2),
                  var3 = rnorm(60, mean = 10, sd = 1),
                  var4 = rnorm(60, mean = 5, sd =2))
data$group<-as.factor(data$group)

continuous_2g_pair(data = data, groupvar = "group")

# Set names to variables
if(requireNamespace("table1")){
table1::label(data$var1) <- "Variable 1"
table1::label(data$var2) <- "Variable 2"
table1::label(data$var3) <- "Variable 3"
table1::label(data$var4) <- "Variable 4"

continuous_2g_pair(data = data, groupvar = "group", flextableformat = FALSE)
}

Bivariate analysis for correlation tests

Description

Automatic correlation analyses for continuous variables with one variable as reference. Variable names can be assigned using table1::label() function.

Usage

continuous_corr_test(
  data,
  referencevar,
  alternative = NULL,
  flextableformat = TRUE,
  corr_test = c("all", "pearson", "spearman", "kendall")
)

Arguments

Value

A dataframe or flextable containing pvalues for correlation tests along with the normality and homocedasticity tests p values

Examples

# example code

data <- data.frame(group = rep(letters[1:2], 15),
var1 = rnorm(30, mean = 15, sd = 5),
var2 = rnorm(30, mean = 20, sd = 2),
var3 = rnorm(30, mean = 10, sd = 1),
var4 = rnorm(30, mean = 5, sd =2))

data$group<-as.factor(data$group)

continuous_corr_test(data = data, referencevar = "var1", flextableformat = FALSE)

# Set names to variables
if(requireNamespace("table1")){
table1::label(data$var2) <- "Variable 2"
table1::label(data$var3) <- "Variable 3"
table1::label(data$var4) <- "Variable 4"

continuous_corr_test(data = data, referencevar = "var1", flextableformat = FALSE)
}

# Example performing correlation test for only one variable
if(requireNamespace("dplyr")){
library(dplyr)
continuous_corr_test(data = data %>% select("var1","var2"),
 referencevar = "var1", flextableformat = FALSE, corr_test = "pearson")
}

# Example performing only pearson correlation test
continuous_corr_test(data = data, referencevar = "var1",
 flextableformat = FALSE, corr_test = "pearson")

Bivariate analysis for more than 2 groups

Description

Generates a HTML table of bivariate analysis for 2 groups.

Usage

continuous_multg(data, groupvar, flextableformat = TRUE)

Arguments

Value

A dataframe or flextable containing pvalues for each test along with the normality and homocedasticity tests p values. An extra column will be shown indicating the recommended significant test

Examples

data <- iris

data$Species<-as.factor(data$Species)

continuous_multg(data = data, groupvar = "Species", flextableformat = FALSE)

Bivariate Chi squared and Fisher Test analysis for 2 categories.

Description

Generates a HTML table of bivariate Chi squared and Fisher Test analysis for 2 categories. Display a table arranged dataframe with Chi squared statistic, minimum expected frecuencies, Chi squared p value, Fisher Test p value, and Odds ratio with 95 confidence levels. Note that you must recode factors and level the database factors in order to compute exact p values. Variable names can be assigned using table1::label() function.

Usage

dichotomous_2k_2sid(data, referencevar, flextableformat = TRUE)

Arguments

Value

Returns a dataframe or flextable containing statistical values for Chi squared tests or Fisher's test.

Author(s)

JAFG

Examples

  # Not run

 # Create a sample dataframe
 df <- data.frame(
   has = c("Yes", "No", "Yes", "Yes", "No", "No", "Yes"),
   smoke = c("Yes", "No", "No", "Yes", "No", "Yes", "No"),
   gender = c("Male", "Female", "Male", "Female", "Female", "Male", "Male"))

 df$has <- as.factor(df$has)
 df$smoke <- as.factor(df$smoke)
 df$gender <- as.factor(df$gender)

# Set a value as reference level
 df$has <- relevel(df$has, ref= "Yes")
 df$smoke <- relevel(df$smoke, ref= "Yes")
 df$gender <- relevel(df$gender, ref= "Female")

 # Apply function
dichotomous_2k_2sid(df, referencevar="has")
dichotomous_2k_2sid(df, referencevar="has", flextableformat = FALSE)

# Set names to variables
if(requireNamespace("table1")){
table1::label(df$has) <- "Hypertension"
table1::label(df$smoke) <- "Smoking Habits"
table1::label(df$gender) <- "Gender"

dichotomous_2k_2sid(df, referencevar="has", flextableformat = FALSE)
}

Encode character variables as factor automatically

Description

Encode character variables as factor automatically

Usage

encode_factors(
  data,
  encode = c("character", "integer"),
  list_factors = NULL,
  uselist = FALSE
)

Arguments

Value

Converts listed columns to factors.

Examples

df <- data.frame(has = c("Yes", "No", "Yes", "Yes", "No", "No", "Yes"),
smoke = c("Yes", "No", "No", "Yes", "No", "Yes", "No"),
gender = c("Male", "Female", "Male", "Female", "Female", "Male", "Male"))

str(df)

df <- encode_factors(df, encode = "character")

str(df)

Summary method for logistf with no printable output

Description

Summary method for logistf models, currently this method is only used in step_bw_firth function.

Usage

logistf_summary(object, verbose = FALSE, ...)

Arguments

Value

An object class 'data.frame' showing coefficients and p_values.

References

Heinze G, Ploner M, Jiricka L, Steiner G. logistf: Firth’s Bias-Reduced Logistic Regression. 2023. available on: https://CRAN.R-project.org/package=logistf

Examples

# Only use if you want a non-printable version of 'summary' for a logistfnp object.
if(requireNamespace("logistf")){
library(logistf)
data <- mtcars
data$am <- as.factor(data$am)

regression_model <- logistf::logistf(am ~ mpg + cyl + disp, data = data)
class(regression_model) <- c("logistfnp")
summary(regression_model)
}

Sample Size Calculation for multiple regression analysis

Description

Calculates the recommended sample size for a multiple regression analysis.

Usage

ss_multreg(df, prop = NULL, logistic = FALSE, verbose = TRUE)

Arguments

Value

An object class ss_multreg_obj indicating the sample size calculation for a regression analysis.

References

Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR. A simulation study of the number of events per variable in logistic regression analysis. Journal of Clinical Epidemiology. diciembre de 1996;49(12):1373–9.

Pierdant-Pérez M, Patiño-López MI, Flores-García JA, Jacques-García FA. Implementación de un curso virtual de lectura crítica en estudiantes de medicina durante la pandemia COVID-19. Inv Ed Med. el 1 de octubre de 2023;12(48):64–71.

Examples

# Lineal multiple regression with 4 degrees of freedom
ss_multreg(4, logistic = FALSE)

# Logistic multiple regression with 4 degrees of freedom
# and 60% of probability of the event

ss_multreg(4, prop = .6, logistic = TRUE)

Stepwise backward for logistic Firth regression with automated dummy variables conversion

Description

Extension code to perform stepwise backward to a logistf model with categorical variables. Automatically transforms predictors of the model which are factors to dummy variables.

Usage

step_bw_firth(
  reg_model,
  s_lower = "~1",
  s_upper = "all",
  trace = TRUE,
  steps = NULL,
  p_threshold = 0.05,
  data = NULL
)

Arguments

Value

An oject class step_bw containing the final model an each step performed in backward regression. The final model can be accessed using $ operator

References

Heinze G, Ploner M, Jiricka L, Steiner G. logistf: Firth’s Bias-Reduced Logistic Regression. 2023. Available on: https://CRAN.R-project.org/package=logistf

Efroymson MA. Multiple regression analysis. In: Ralston A, Wilf HS, editors. Mathematical methods for digital computers. New York: Wiley; 1960.

Ullmann T, Heinze G, Hafermann L, Schilhart-Wallisch C, Dunkler D, et al. (2024) Evaluating variable selection methods for multivariable regression models: A simulation study protocol. PLOS ONE 19(8): e0308543

Examples

if(requireNamespace("logistf")){
library(logistf)

data<-mtcars
data$am<-as.factor(data$am)

regression_model<-logistf::logistf(am~mpg+cyl+disp, data=data)
stepwise<-step_bw_firth(regression_model, trace=FALSE)

final_stepwise_model<-stepwise$final_model

# Show steps
stepwise$steps

summary(final_stepwise_model)
}

Automatized stepwise backward for regression models

Description

Automatized stepwise backward for regression models

Usage

step_bw_p(
  reg_model,
  s_lower = "~1",
  s_upper = "all",
  trace = TRUE,
  steps = NULL,
  p_threshold = 0.05,
  data = NULL,
  ...
)

Arguments

Value

An oject class step_bw containing the final model an each step performed in backward regression. The final model can be accessed using $ operator

References

Efroymson MA. Multiple regression analysis. In: Ralston A, Wilf HS, editors. Mathematical methods for digital computers. New York: Wiley; 1960.

Examples

data(mtcars)
regression_model<-lm(cyl~., data=mtcars)
stepwise<-step_bw_p(regression_model, trace=FALSE)

final_stepwise_model<-stepwise$final_model

summary(final_stepwise_model)

Basic theme for Bivaran packages plots

Description

Basic theme for Bivaran packages plots

Usage

theme_serene(
  base_size = 14,
  base_family = "sans",
  base_fontface = "plain",
  base_line_size = base_size/14,
  base_rect_size = base_size/14,
  axis_text_angle = 0,
  border = FALSE
)

Arguments

Value

Returns a list of classes "gg" and "theme"

Author(s)

Jhoselin Marian Castro-Rodriguez

Examples

library(ggplot2)
data <- mtcars
p1 <- ggplot(data, aes(disp, hp))+
geom_point()+
geom_smooth()

p1 + theme_serene()

Void theme for Bivaran packages plots

Description

Basic theme for Bivaran packages plots

Usage

theme_serene_void(
  base_size = 11,
  base_family = "sans",
  base_fontface = "plain",
  base_line_size = base_size/22,
  base_rect_size = base_size/2,
  axis_text_angle = 0,
  border = FALSE
)

Arguments

Value

Returns a list of classes "gg" and "theme"

Author(s)

Jhoselin Marian Castro-Rodriguez

Examples

library(ggplot2)

data <- mtcars
p1 <- ggplot(data, aes(disp, hp))+
geom_point()+
geom_smooth()

p1 + theme_serene_void()