Future of Tuned Ratio Unbiased Mean Predictor (TRUMP) with the Unified Scrambling Approach (USA)

The Tuned Ratio Unbiased Mean Predictor (TRUMP) was introduced by Singh and Sedory (2017: Survey Research Methods Section, Proceedings of the American Statistical Association, pp. 1746-1759). They have shown that the proposed TRUMP when utilizing First Basic Information (FBI) about the TRUMP Care coefficient can perform better than the Best Linear Unbiased Estimator (BLUE) and, also can perform better than the Best Linear Unbiased Predictor (BLUP). Warner (1965: Journal of the American Statistical Association, pp. 63-69) introduced the idea of estimating the population proportion of a sensitive attribute by making use of randomization device. Later on, the idea was extended to estimate the population mean of a sensitive variable by making use of an approach involving additive and multiplicative scrambling variables. In this paper, we will study the future of the TRUMP with a Unified Scrambling Approach (USA) along the lines of Singh, Joarder and King (1996: Australian Journal of Statistics, pp. 201-211). Making a great adjustment (MAGA) by means of scrambling variables may help TRUMP have more precise estimates of frauds, induced abortions, illegal immigration, extramarital relations, tax returns, illegal drugs, and cheating etc. The results based on theory and simulation study will be reported.