Improving quarterly index of turnover by means of a calibration estimator

The sampling design strategies used for estimation of economic indicators in short-term statistics domain are heterogeneous depending on the characteristic the variable to be estimated and timeliness required for the indices. On the occasion of change of the base year for the turnover indices for services sector it has been analysed how to improve in terms of variability the classical Horwitz-Thompson estimator for the estimation of totals using a calibration estimator. With this aim it has been necessary to carry out a simulation experiment to compare the performance of the estimators for different choice of the auxiliary variable used in the calibration estimator. The simulation has been studied to reproduce the characteristic of economic variable and the typical behaviour of short-term indicators.