Asymptotic efficiency of the calibration estimator in a high-dimensional data setting

In a finite population sampling survey, auxiliary information is commonly used to improve the Horvitz–Thompson estimators and calibration has been extensively by national statistical agencies over last decades for that purpose. This method enables make consistent with known totals of variables reduce variance if are explanatory variable interest. Nowadays, it not unusual anymore have high-dimensional data sets adding too much additional may increase estimators. We study in this paper asymptotic efficiency estimator we prove suffer from an variability be neglected certain conditions. suggest bootstrap criterion choice variables. A short simulation shows proposed lead more parsimonious number associated weights smaller variation no inflation.