Nonparametric Generation of Synthetic Data Using Copulas

This article presents a novel nonparametric approach to generate synthetic data using copulas, which are functions that explain the dependency structure of real data. The proposed method addresses several challenges faced by existing generation techniques, such as preservation complex multivariate structures presented in By all information from and verifying generated follows same behavior under homogeneity tests, our is significant improvement over techniques. Our easy implement interpret, making it valuable tool for solving class imbalance problems machine learning models, improving generalization capabilities deep anonymizing finance healthcare domains, among other applications.