In this paper, we develop a methodology for constructing a macro-econometric model. First we combine economic theory and time series techniques. This allows us to adapt the cointegration regression to set up the single equations. Then we use the Gauss-Seidel method to solve the simultaneous equations. We also demonstrate how to link sectors in the model and the procedure for building the model system. We begin with the simplest model and add more relations gradually. Finally, we use a macro-econometric model to analyze the economy in Taiwan, and simulate changes in the fiscal policy, crude oil price and growth rate of mainland China to do scenario analysis.