Optimizing Sensitive Variable Estimation Through Additive Scrambling and Log-Based Techniques

The study introduces a new set of log-type estimators using the additive scrambling model to improve the accuracy of mean estimation for sensitive variables. The Randomized Response Technique protects respondents’ privacy and improves data validity and reliability. The study uses three datasets to evaluate the performance of the new estimators, demonstrating their superior efficiency and reliability. The improved accuracy of the estimators can lead to more reliable statistical analyses and better-informed decision-making processes. The method’s applicability across different datasets and contexts highlights its robustness and utility in various practical scenarios. This research explores cybersecurity and marketing applications and combines these methods with artificial intelligence and machine learning models. Comparative studies with other estimation techniques can help refine the proposed methods and identify areas for improvement. An empirical investigation employing three datasets reveals that the proposed estimators outperform those reported in prior research.