A method of investigating the behaviour of economic models which are too complicated for analytical solutions to be possible. A system is started off at a large number of initial positions chosen at random, and followed through a numerical simulation to see how it evolves. Monte Carlo methods can be used to check whether a system has an equilibrium, and whether this is stable for any starting point, or for some limited region of possible starting points. Monte Carlo simulations are used in econometrics to approximate the true sampling properties of an estimator of a model parameter or of a test statistic. This is achieved by generating a large number of independent artificial data sets under the conditions of interest and calculating the parameter estimate or the statistic for each data set. For example, the sample mean of the parameter estimate across the data sets is used to approximate the true mean of the sampling distribution of the correspondent estimator.