Design and implementation of a system to determine tax evasion through de stochastic techniques

The dominant view on the study and practice of the tax service, mentions that taxpayers will always seek their own interest, for that reason taxes will be considered costs that could be avoided, unless the probability of being detected is high and the severity of the penalty is not an attractive option. This means that taxpayers would be motivated to develop tax evasion strategies that would increase their economic benefits. In this regard, the following two types of intentional tax evasion procedures are identified: one, where illegal practices are employed, such as failing to file a tax return without any legal justification to do so, and, another, where legal procedures are used to avoid filing a tax return. This paper establishes a methodology based on stochastic techniques, with the objective of creating a model that allows the identification of possible tax evaders of the value added tax (VAT) in Colombia.