EVALUATING THE STRENGTH OF IDENTIFICATION IN DSGE MODELS. AN A PRIORI APPROACH

摘要: This paper presents a new approach to parameter identification analysis in DSGE models wherein the strength of is treated as property underlying model and studied prior estimation. The reflects empirical importance economic features represented by parameters. Identification problems arise when some parameters are either nearly irrelevant or redundant with respect aspects reality designed explain. therefore not only crucial for estimation models, but also has important implications development. proposed measure based on Fisher information matrix depends three factors: values, set observed variables sample size. By applying methodology, researchers can determine effect each factor individual parameters, study how it related structural statistical characteristics model. methodology illustrated using medium-scale estimated Smets Wouters (2007).