We discuss the use of calibration techniques in economic models. Calibration contrasts with estimation in relying on deterministic calculation of model parameter values consistent with data, rather than econometric estimation. The reasons why calibrators use these methods, as well as the main arguments in debates between calibrators and econometricians are set out. We draw a distinction between the calibration methods used in dynamic macro models of the Kydland-Prescott type and micro models of the Shoven-Whalley variety. We highlight the ways in which calibration techniques are evolving including double calibration, the use of data pre-adjustments, and the incorporation of model estimation consistent elasticities. We conclude with a discussion of what constitutes best practice in calibration.
