I propose a method of endogenous calibration of nonlinear stochastic equilibrium models, by taking advantage of spectral properties of economic data. Model parameters are determined by minimizing a distance metric specified in frequency domain. Several weight functions are discussed and their usefulness is assessed. Both univariate and multivariate settings are considered. Models' statistical testing strategies are discussed.
|Original language||American English|
|Title of host publication||WCGJ International Conference Commemorating the 55th Anniversary of the State of Israel|
|State||Published - 2003|