Numerical methods in probability theory can generally be divided into two groups, deterministic and Monte Carlo methods. Deterministic methods are used for numerical expectations of various random variables whenever such methods are efficient from the point of view of the time spent and the complexity of the numerical procedure. It is widely believed that Monte Carlo methods are used exclusively as an alternative numerical procedure for the numerical estimation of the expectation of random variables. At the same time, it is considered that Monte Carlo methods should be used when they are more efficient than deterministic ones. However, there are numerical procedures in which Monte Carlo methods are an indispensable tool. It is enough to mention statistical models in technical, natural and social sciences, and the bootstrap method for estimating the parameters of a statistical model. Therefore, Monte Carlo methods must be seen as an essential tool in the numerical procedures of probability theory, and should be known as an integral part of probability theory.