Radial basis functions neural networks for nonlinear time series analysis and time-varying effects of supply shocks

I propose a flexible Radial Basis Functions (RBFs) Artificial Neural Networks method for studying the time series properties of macroeconomic variables. To assess the validity of the RBF approach, I conduct a Monte Carlo experiment using the data generated from a nonlinear New Keynesian (NK) model. I find that the RBF estimator can uncover the structure of the NK model from the simulated data of 300 observations. Finally, I apply the RBF estimator to the quarterly US data and show that the positive supply shocks have significantly weaker expansionary effects during the periods of passive monetary policy regimes.