Automated Exploratory Data Analysis — autoeda • ClinicoPath

Automated Exploratory Data Analysis using DataExplorer package integration. This module provides comprehensive automated EDA capabilities including dataset overview, missing value analysis, correlation matrices, PCA visualization, and automated reporting. Based on autoEDA research from R Journal 2019 (Staniak & Biecek).

Usage

autoeda(
  data,
  vars,
  analysis_type = "overview",
  target_var,
  include_plots = TRUE,
  missing_threshold = 5,
  correlation_method = "pearson",
  pca_components = 5,
  plot_theme = "clinical",
  output_format = "combined",
  eda_engine = "dataexplorer",
  advanced_options = FALSE,
  categorical_limit = 15,
  generate_report = FALSE
)

Arguments

data: The data as a data frame.
vars: Variables to include in the automated exploratory data analysis. All variable types are supported.
analysis_type: Choose the type of automated EDA analysis to perform.
target_var: Optional target variable for supervised EDA analysis. Used for target vs predictors analysis.
include_plots: Include automated plots and visualizations in the output.
missing_threshold: Threshold percentage for highlighting variables with missing values.
correlation_method: Method for correlation analysis.
pca_components: Number of principal components to display in PCA analysis.
plot_theme: Visual theme for automated plots.
output_format: Format for the automated EDA output.
eda_engine: Choose the exploratory data analysis engine. DataExplorer provides comprehensive automated reporting. ggEDA provides enhanced visualizations for clinical research.
advanced_options: Enable advanced features and detailed analysis.
categorical_limit: Maximum number of levels for categorical variables to include in analysis.
generate_report: Generate a comprehensive automated EDA report with all analyses.

Value

A results object containing:

`results$todo`					Welcome message and instructions
`results$overview`					Comprehensive dataset introduction and summary
`results$missing_analysis`					Missing value patterns and recommendations
`results$distributions`					Univariate distribution analysis and plots
`results$correlation_analysis`					Correlation matrices and relationship analysis
`results$pca_analysis`					PCA results and component visualization
`results$target_analysis`					Target vs predictors relationship analysis
`results$comprehensive_report`					Complete automated exploratory data analysis report
`results$plots`					Collection of automated plots and charts
`results$recommendations`					Automated recommendations for further analysis

Examples

# \donttest{
# Example:
# 1. Load your data frame.
# 2. Select variables for analysis.
# 3. Choose analysis type (overview, missing, correlation, etc.)
# 4. Configure output options.
# 5. Run autoeda module for comprehensive automated analysis.
# }