Abstract:Regression analysis is one of the most popularly used statistical technique which only measures the direct effect of independent variables on dependent variable. Path analysis looks for both direct and indirect effects of independent variables and may overcome several hurdles allied with regression models. It utilizes one or more structural regression equations in the model which are used to estimate the unknown parameters. The aim of this work is to study the path analysis models when the endogenous (dependent) variable and exogenous (independent) variables are linked through the elliptical copulas. Using well-organized numerical schemes, we investigate the performance of path models when direct and indirect effects are estimated applying classical ordinary least squares and copula-based regression approaches in different scenarios. Finally, two real data applications are also presented to demonstrate the performance of path analysis using copula approach.

What problem does this paper attempt to address?

This paper attempts to address the problem of estimating causal effects in path analysis models when endogenous (dependent) variables and exogenous (independent) variables are correlated through an elliptical Copula. Specifically, the paper explores the performance of using classical least squares and Copula-based regression methods to estimate direct and indirect effects in different scenarios, and validates the effectiveness of these methods through numerical simulations and real data applications. ### Main Objectives of the Paper: 1. **Study Path Analysis Models**: Investigate the performance of path analysis models when endogenous and exogenous variables are correlated through an elliptical Copula. 2. **Compare Different Methods**: Compare the performance of classical least squares and Copula-based regression methods in estimating direct and indirect effects. 3. **Validate with Real Applications**: Demonstrate the superiority of Copula-based path analysis methods through real data applications. ### Key Points of the Paper: - **Path Analysis**: Extends multiple regression to examine causal relationships among multiple variables, including both direct and indirect effects. - **Elliptical Copula**: Links endogenous and exogenous variables through an elliptical Copula, addressing issues such as non-normality and multicollinearity. - **Numerical Simulation**: Evaluates the performance of different methods through simulations of three-variable and four-variable path models with different sample sizes and correlations. - **Real Data Applications**: Uses two real data sets to further validate the effectiveness of Copula-based path analysis methods. ### Conclusion: - **Copula-based Regression Methods**: Copula-based regression methods are more robust in estimating direct and indirect effects, especially when dealing with non-normality and multicollinearity issues. - **Real Applications**: Real data applications further demonstrate the effectiveness and superiority of Copula-based path analysis methods.

Copula-Based Estimation of Causal Effects in Multiple Linear and Path Analysis Models

Parametric and semiparametric approaches for copula-based regression estimation

Copula-Based Regression with Mixed Covariates

Robustness of copula-correction models in causal analysis: Exploiting between-regressor correlation

A Novel Nonparametric Estimation For Conditional Copula Functions Based On Bayes Theorem

Structural and Predictive Analyses with a Mixed Copula‐based Vector Autoregression Model

Estimating complex causal effects from incomplete observational data

A New Copula Model-Based Method for Regression Analysis of Dependent Current Status Data

Common Correlated Effects Estimation of Nonlinear Panel Data Models

Testing Model Fit in Path Models with Dependent Errors Given Non-Normality, Non-Linearity and Hierarchical Data

A Functional Principal Component Analysis Approach to Conditional Copula Estimation

Copula-based analysis of dependent current status data with semiparametric linear transformation model

PATH ANALYSIS WITH LOGISTIC REGRESSION MODELS : EFFECT ANALYSIS OF FULLY RECURSIVE CAUSAL SYSTEMS OF CATEGORICAL VARIABLES

On the Use of Two-Way Fixed Effects Regression Models for Causal Inference with Panel Data

Efficient Algorithms for Covariate Analysis with Dynamic Data Using Nonlinear Mixed-Effects Model.

Maximum Likelihood Estimation Methods for Copula Models

An R package for parametric estimation of causal effects

A copula approach for dependence modeling in multivariate nonparametric time series

The causal interpretation of estimated associations in regression models

Improving forecasting performance using covariate-dependent copula models

A Copula-Based Portrayal of the Collider Bias