The purpose of this project is to develop techniques of nonlinear time series analysis and apply these techniques to answer a number of interesting questions arising from both aggregate and disaggregate economic time series. There are two general types of techniques that need to be developed: first, methods of analyzing properties of multivariate nonlinear time series models and second, methods to detect and identify nonlinear structure in economic time series. The econometric tools developed in this project are applied to a number of issues where linear time series models would be inadequate. These issues include: (1) evaluating the capability of a menu cost pricing and oligopoly model to explain the price dynamics in a large panel of wholesale gasoline rack prices and (2) examining how monetary policy affects the economy by using nonlinear vector autoregression techniques on aggregate output and on more disaggregate measures. For example, are smaller firms more sensitive to changes in monetary policy during recessions than expansions?
