In the previous chapter, we were concerned almost exclusively with the problem of how to generate independent random variables with a specified distribution. In the simplest settings, this is the underlying statistical model and we need go no further. In many other situations, we have to simulate some sort of dependent data or noise process to act as inputs to our simulation model. Many dependent stochastic models can be simulated in an obvious way from their definitions. Nevertheless, some tricks sometimes are useful and we present a few of the more common ones below.
KeywordsMarkov Chain Discrete Fourier Transform ARMA Model Metropolis Algorithm Gaussian Random Process
Unable to display preview. Download preview PDF.