Comparing Methods for Solving General Equilibrium Models with Incomplete Markets and Portfolio Choice

In this paper, I compare the accuracy of the two existing methods for solving stochastic general equilibrium models with dynamic portfolio choice and incomplete markets: one proposed by Hnatkovska (2010) and Evans and Hnatkovska (2005, 2011) (EH), another - attributed to Devereux and Sutherland (2007, 2009a,b) together with Tille and van Wincoop (2007) (DS). The accuracy of these solution methods for the real as well as portfolio variables is analyzed by studying the distribution of Euler equation errors and using a series of accuracy tests. My results indicate that while both DS and EH methods generate su ciently accurate solutions for the real variables, there are significant gains from using the EH method when solving for the portfolio solutions. This implies that both the DS and EH methods can be used to analyze the impulse responses of any variables, other than portfolio variables, or to investigate the business cycle properties of a model. If, on the other hand, a researcher is interested in analyzing the dynamics of the portfolio variables, an extra care should be taken to assess the accuracy of the solution, especially that of the portfolio side of the model