Hi It generally means choosing the model's parameters to get the best possible agreement between the model's outputs and observed data. So, for example, you might check the callibration of an asset pricing model by seeing how the model's output prices compare with current observed asset market prices. Lynn
