Resumen de Calibrating a household survey by using the CALMAR program
This paper addresses the problem of calibrating a survey to some external information. In the rst part of the paper, after introducing the key concepts and basic de nitions, the calibration techniques are explained. In the second part, a simulation experiment is performed: survey samples are taken from a census population, non-response is also simulated and the basic (Horvitz-Thompson) design estimator of some population totals is compared with a ratio estimator and with a calibration estimator. The conclusions are that calibration estimates are in general more accurate than those of other kind of estimators.
