The generalized method of moments, often abbreviated as GMM, is a method of solving a system of equations where exogenous variables are correlated with error terms and the probability distribution of data conditional on model parameters is intractable to compute. The method can be used for linear or non-linear system of equations. In the context of linear equations, we observed in an earlier chapter that the existence of correlation between exogenous variables and error term causes parameter estimates from OLS to be biased and inconsistent. This necessitated the use of instrumental variables. These variables are dependent on exogenous variables and uncorrelated with error terms. In the context of linear system of equations, GMM can be viewed as a generalization of the method of instrumental variables for estimating model parameters.