ana: R Package for Data Analysis and Visualization

Overview

The ana package provides a comprehensive toolkit for exploratory data analysis in R, featuring automatic variable analysis, visualization, and survival time calculations. The package includes both Chinese interface functions and optimized survival analysis utilities.

Installation

Common Installation Issues

Many users encounter the error Error in library(ana) : there is no package called 'ana' after installing from GitHub. This typically occurs because:

1. The package lacks proper R package structure
2. Missing DESCRIPTION file
3. No proper namespace configuration

Proper Installation Methods

Method 1: Source Installation (Recommended)

# Download the repository
# Then source the main file directly:
source("path/to/ana.R")

# For survival time functions, source individually as needed:
source("path/to/a_survival_time.R")
source("path/to/a_survival_time_table.R")
# etc.

Method 2: Create Local Package Structure

Create a proper package structure first:

ana/
├── DESCRIPTION
├── NAMESPACE
├── R/
│   ├── ana.R
│   ├── a_survival_time.R
│   ├── a_survival_time_table.R
│   ├── a_survival_time_C++.R
│   ├── a_survival_time_large.R
│   └── a_survival_time_ram.R
└── man/

Create DESCRIPTION file:

Package: ana
Type: Package
Title: Data Analysis and Visualization Tools
Version: 1.0.0
Author: Your Name
Maintainer: Your Name <your.email@example.com>
Description: Comprehensive toolkit for exploratory data analysis with automatic 
    variable analysis, visualization, and survival time calculations.
License: GPL-3
Encoding: UTF-8
LazyData: true
Imports:
    haven,
    ggplot2,
    dplyr,
    scales,
    knitr,
    rmarkdown,
    data.table,
    parallel,
    foreach,
    doParallel,
    Rcpp
RoxygenNote: 7.2.3

Create NAMESPACE file:

export(ana)
export(alook)
export(avar)
export(a_survival_time_improved)
export(a_survival_time_dt)
export(a_survival_time_rcpp)
export(a_survival_time_parallel)
export(a_survival_time_memory)

Then install:

# Install from local directory
install.packages("path/to/ana", repos = NULL, type = "source")

# Or use devtools
devtools::install("path/to/ana")

Dependencies

The package requires the following R packages:

Core Dependencies

• haven: For reading SPSS/Stata/SAS files
• ggplot2: For visualization
• dplyr: For data manipulation
• scales: For plot scales
• knitr: For reporting
• rmarkdown: For document generation

Additional Dependencies for Survival Functions

• data.table: For optimized data.table version
• Rcpp: For C++ optimized version
• parallel, foreach, doParallel: For parallel processing version

Install all dependencies:

install.packages(c("haven", "ggplot2", "dplyr", "scales", 
                   "knitr", "rmarkdown", "data.table", 
                   "Rcpp", "parallel", "foreach", "doParallel"))

Main Functions

1. ana() - Basic Analysis

Performs comprehensive descriptive statistics on specified variables.

# Analyze specific variables
ana(mydata, "var1", "var2", "var3")

# Analyze all suitable variables (auto-selects up to 10)
ana(mydata)

Features:

• Automatic type detection (numeric, factor, character, logical, date)
• Missing value analysis
• Descriptive statistics for numeric variables
• Frequency tables for categorical variables
• Handles special values (Inf, NaN)

2. alook() - Visual Analysis

Creates automatic visualizations based on variable types.

# Visualize specific variables
alook(mydata, "var1", "var2")

# Visualize all suitable variables
alook(mydata)

Visualizations:

• Missing value bar charts
• Histograms for numeric variables
• Box plots with jitter for small datasets
• Bar charts for categorical variables

3. avar() - All Variables Analysis

Provides a comprehensive overview of all variables in the dataset.

avar(mydata)

Output includes:

• Variable type classification
• Missing value summary
• Data quality score
• Single-value variable detection
• Data quality recommendations

Survival Time Analysis Functions

The package includes multiple optimized versions of survival time calculation:

1. a_survival_time_improved() - Enhanced Base Version

The main survival time calculation function with comprehensive error handling.

# Basic usage
result <- a_survival_time_improved(data, 
                                   id_var = "patient_id",
                                   time = "visit_time", 
                                   event = "event_occurred",
                                   check_order = TRUE,
                                   verbose = TRUE)

Parameters:

• data: Data frame containing the survival data
• id_var: Column name for individual/subject ID
• time: Column name for time variable
• event: Column name for event indicator (0/1)
• check_order: Check temporal ordering of observations
• verbose: Print warnings and messages
• handle_negative_time: Handle negative time values

2. a_survival_time_dt() - data.table Optimized Version

5-10x faster for large datasets using data.table.

# Requires data.table package
library(data.table)
result <- a_survival_time_dt(data, 
                             id_var = "patient_id",
                             time = "visit_time", 
                             event = "event_occurred")

3. a_survival_time_rcpp() - C++ Optimized Version

10-20x faster using Rcpp for maximum performance.

# Requires Rcpp package
library(Rcpp)
result <- a_survival_time_rcpp(data, 
                               id_var = "patient_id",
                               time = "visit_time", 
                               event = "event_occurred")

4. a_survival_time_parallel() - Parallel Processing Version

For very large datasets, uses multiple CPU cores.

# Automatically detects cores
result <- a_survival_time_parallel(data, 
                                   id_var = "patient_id",
                                   time = "visit_time", 
                                   event = "event_occurred",
                                   n_cores = 4)  # Optional

5. a_survival_time_memory() - Memory Optimized Version

For memory-constrained environments, uses optimized data types.

result <- a_survival_time_memory(data, 
                                 id_var = "patient_id",
                                 time = "visit_time", 
                                 event = "event_occurred")

Usage Examples

Example 1: Basic Data Analysis

# Load sample data
data(iris)

# Basic analysis of specific variables
ana(iris, "Sepal.Length", "Species")

# Visual analysis
alook(iris, "Sepal.Length", "Species")

# Analyze all variables
avar(iris)

Example 2: Survival Analysis

# Create sample survival data
survival_data <- data.frame(
  patient_id = rep(1:100, each = 5),
  visit_time = rep(0:4, 100) + runif(500, -0.1, 0.1),
  event = rbinom(500, 1, 0.1)
)

# Calculate survival times
result <- a_survival_time_improved(survival_data,
                                   id_var = "patient_id",
                                   time = "visit_time",
                                   event = "event")

# For large datasets, use optimized versions
result_fast <- a_survival_time_dt(survival_data,
                                  id_var = "patient_id",
                                  time = "visit_time",
                                  event = "event")

Troubleshooting

Issue: Package not found after GitHub installation

Solution 1: Use source() instead of library()

source("ana.R")
# Now functions are available directly
ana(mydata)

Solution 2: Create proper package structure (see Installation section)

Issue: Dependencies not automatically installed

Solution: Manually install all dependencies

# Check and install missing packages
required_packages <- c("haven", "ggplot2", "dplyr", "scales", 
                      "knitr", "rmarkdown", "data.table", "Rcpp")
missing_packages <- required_packages[!required_packages %in% installed.packages()[,"Package"]]
if(length(missing_packages) > 0) {
  install.packages(missing_packages)
}

Issue: Rcpp functions not working

Solution: Ensure Rcpp is properly installed and configured

# Test Rcpp
install.packages("Rcpp")
library(Rcpp)
evalCpp("2 + 2")  # Should return 4

Issue: Parallel functions not working on Windows

Solution: Windows requires special handling for parallel processing

# On Windows, use:
library(parallel)
cl <- makeCluster(detectCores() - 1)
# ... your code ...
stopCluster(cl)

Performance Comparison

Function	Relative Speed	Best Use Case
a_survival_time_improved	1x (baseline)	Small datasets, full features
a_survival_time_dt	5-10x	Medium to large datasets
a_survival_time_rcpp	10-20x	Large datasets, speed critical
a_survival_time_parallel	Varies	Very large datasets, multi-core systems
a_survival_time_memory	0.8x	Memory-constrained environments

Contributing

Feel free to submit issues, fork the repository, and create pull requests for any improvements.

License

GPL-3

Notes

• The package automatically loads required dependencies on attach
• Chinese characters in function output may require proper encoding settings
• For best performance with large datasets, use the optimized survival time functions
• The visualization functions (alook) work best with datasets under 10,000 rows

Citation

If you use this package in your research, please cite:

@Manual{,
  title = {ana: Data Analysis and Visualization Tools for R},
  author = {Author Name},
  year = {2024},
  note = {R package version 1.0.0},
  url = {https://github.com/username/ana},
}