Before you begin the classification of risks, it is worth giving a definition to this concept.

Thus, the risk can be called a possible undesirable event, or a list of undesirable events, during which the control object goes into a certain undesirable state.

This condition is called a risk situation. In this case, by the realization of risk we will understand the occurrence of a risk event and the transition of the control object to this undesirable state under the influence of this event.

To date, there is no established classification of risks, perhaps this is due to an increase in the varieties of risks both in the financial sector and in other sectors. In this monograph, a methodology is used to divide risks into certain subgroups, which was proposed by Professor A. I. Orlov [3].

The totality of economic risks should be divided into groups, where the movement from the particular to the general is taken as the ordering principle. According to this principle, these risk classes can be distinguished:

- production risks. The risks under consideration are directly related to the activities of an enterprise or organization;

- financial risks. These risk classes are related to the macroeconomic situation;

- commercial risks. It is worth noting that these risks may be caused by incomplete predictability of market dynamics;

- risks that arise at the level of the state or the Earth as a whole.

Next, we consider specifically what risks relate to melon risk classes. Production risks include: risks of disruption of production processes; risks for the release of defective products or goods; risks of errors in product design; social risks.

Financial risks include: risks associated with inflation; currency risks; risks associated with investments; all kinds of legislative risks.

Commercial risks include: risk associated with the loss of suppliers; various information risks; risk caused by the variability of consumer preferences; regional risks that are caused by the state of individual regions.

The fourth type of risk also plays a huge role, so there is no one state that would not have this type of risk.

The risks that arise at the level of the state or the Earth as a whole should include: external economic risks; industry risks; political risks; as well as risks associated with natural phenomena [4].

 Of all the existing classification methods, the risk classification system of professor A. I. Orlova is the most practical and convenient. It is worth noting that the theory of risk management, which also includes insurance and risk management, is aimed at determining the sources of losses, also studies the logic and probability of occurrence of risk events, and also determines the mechanisms for compensating for the losses associated with them.

The main mechanism for managing risks and losses from risks is hedging. «Hedging is a dynamic strategy for managing an object subject to risks, which provides, with a given degree of accuracy, a quantitative assessment of the possibility of the management object falling into a risky situation and restricts the size of potential losses to a given level when a risk is realized» [4].

According to the analysis of various methods and means of measuring and managing financial risks, it can be concluded that since the beginning of the nineties there has been a massive application in practice of statistical models for assessing market risk losses VAR (Value At Risk) [5–7].

According to the analysis, today for the purpose of computer and mathematical modeling for the description of uncertainties apply:

- conflict theory methods;

- probabilistic-statistical methods;

- methods of statistics of non-numerical data, also methods of the theory of fuzziness;

- methods of the theory of artificial intelligence (neural networks, genetic algorithms, etc.). It should also be noted here that formal methods for assessing risks often do not provide unambiguous conclusions. Therefore, together take off should apply the methods of expert forecasting and the corresponding methods from the theory of expert estimates [3].

It is worth noting that crises at the end of the nineties showed certain limits of applicability of statistical risk models, when it was possible to observe the bankruptcy of some large financial organizations and enterprises [7].

Today, risk management methods based on the use of fuzzy logic and the apparatus of neural networks are widely used. In the following chapters, we will consider the advantages and disadvantages of the currently used econometric models.

It is also worth noting that the disadvantages of existing approaches to risk analysis include the lack of a methodological basis for an integrated analysis of qualitative and quantitative risk factors.

The resolution of the contradictions of the existing methods of analysis and risk assessment can be obtained by applying fuzzy models [8].