Network Calibration and Metamodeling of a Financial Accelerator Agent Based Model

We introduce a financially constrained production framework in which heterogeneous firms and banks entertain multiple credit connections. The parameters of credit market interaction are estimated from real data in order to reproduce a set of empirical regularities of the Japanese credit market. Then, we pursue the metamodeling approach, i. e. we derive a reduced form for a set of simulated moments through the following steps: 1. we run agent-based simulations using an efficient sampling design of the parameter space; 2. we employ the simulated data to estimate and then compare a number of alternative statistical metamodels. Then, using the most fitting metamodels, we study through sensitivity analysis the effects on the moments of variations of the parameters. Finally, we employ the same approach to calibrate our agent-based model (ABM) with Japanese data, showing that metamodels can provide a clear-cut answer to the question "does the ABM replicate empirical data?"