Regularized Hybrid SEM Framework: R Code and Simulated Data for Market Segmentation Analysis

The framework integrates three methodological approaches:Framework Components:Stage 1: Reflective Measurement - Confirmatory Factor Analysis (CFA) using lavaan packageStage 2: Formative Composites - Elastic Net regularization using glmnet packageStage 3: Heterogeneity Modeling - Latent Profile Analysis (LPA) using mclust packageInference - Full-information bootstrap with Bias-Corrected and Accelerated (BCa) confidence intervalsIncluded Files:hybrid_sem_main.R - Core analysis functionshybrid_sem_methods.R - S3 methods for results objectshybrid_sem_demo.R - Demonstration and utility functionsREADME.md - Complete documentation and usage instructionssimulated_data.csv - Example dataset (n=450) for demonstrationexample_analysis.R - Complete analysis exampleSimulated Data Characteristics:Sample size: 450 observationsVariables:5 reflective indicators (Likert-scale, 1-7)6 formative indicators (behavioral measures)1 outcome variable (brand advocacy)True segment membership (for validation)Three latent segments with different relationship patternsApplications:Market segmentation with psychological profilesCustomer journey analysisBrand relationship modelingConsumer behavior researchService quality evaluationTechnical Requirements:R version ≥ 4.0.0Required packages: lavaan, glmnet, mclust, boot, dplyr, MASS, parallelMinimum sample size: 200-300 observations for stable estimation