Actually, the methods of forecasting include four methods of working with experts. These are the methods of correspondence and correspondence surveys, the method of commission and the method of delphi. All other methods of working with experts are designed to solve problems of a different class and can be used for forecasting only in individual cases.
The method of correspondence questionnaires consists in the fact that experts are sent questionnaires with questions and a request to respond by such a deadline. According to the above classification, this is a correspondence individual survey. Its advantages include the lowest labor intensity, the disadvantages – less accuracy, since each expert makes forecasting only on the basis of his idea of the object of forecasting and does not take into account the diversity of information available to other experts. In addition, there is always a danger that some experts will misunderstand the question and the organizers of the survey do not have the opportunity to identify this.
The interview method consists in the fact that the organizer of the forecast consistently meets with all experts and personally interviews them. It’s a personal individual method. In this case, the labor intensity increases somewhat (it is necessary to meet with each expert), but it becomes possible to eliminate the problem of a misunderstood question and the accuracy due to this increases.
In the method of the commission, experts carry out forecasting, jointly discussing the problem. It’s a face-to-face collective method. For forecasting by the method of the commission, it is necessary to organize a meeting of all experts, for busy people this is usually far from easy to implement. But as a result, the accuracy of forecasting increases since there is no problem of a misunderstood issue (in the process of discussion, a misunderstanding of the essence of the issue under discussion is easily revealed) and most importantly, experts have the opportunity to take advantage of the whole variety of information about the object of forecasting since in the process of discussion, experts express different points of view. At the same time, there are a number of negative aspects in this method, mainly of a psychological nature, which do not allow to fully realize all the possibilities for improving the accuracy of forecasting. These points include the effect of the psychological pressure of the opinion of the majority and the opinion of the authorities on the individual opinion of the expert. In addition, to hear new information about the object of forecasting prevents the effect of the first word. It is as follows: in an effort to protect their competence in the eyes of others, experts tend to defend a previously publicly expressed point of view, which means that they are psychologically less likely to perceive others who differ from their point of view, after they have publicly expressed their own.
Unlike the three previous forecasting methods, which have both strengths and weaknesses, the fourth forecasting method – the Delphi method – combines the advantages of all its predecessors and is also free from most of their shortcomings. It was developed by the famous consulting firm “Rand Corporation” and named after the mythical Delphic oracle – a soothsayer, who faithfully answered any questions of the inhabitants of ancient Greece through the mouths of the gods.
It refers to the methods of collective correspondence survey of experts and consists of the following stages.
Experts are sent questions. For the answers received, there are median, first and third quartiles. All experts from the first and last quarters are asked to provide arguments in support of their forecasts. The significance of the median, as well as the anonymous arguments of the experts whose forecasts were in the first and last quarters, are sent to all experts with a request to make another forecast in the light of this additional information. For the replies received, the median and the first and third quartiles are again located. If the expert opinion remains unconvinced, steps 3 to 5 are repeated until the required degree of consistency is obtained.
The advantages of the method include low labor intensity, due to the correspondence form of the survey, and high accuracy achieved due to the following factors. Thanks to the arguments of the experts who gave the highest and lowest marks, each expert participating in the next round of forecasting uses all the information about the object of forecasting known to the entire community of experts. Thanks to the anonymity of arguments and ignorance of the answers of specific experts, the pressure of authorities and the effect of adherence to the first word are eliminated. The only negative effect that has not been completely eliminated in this method of forecasting is the effect of the pressure of the majority opinion that with the repeated repetition of points 3-5, there is a tendency to artificially “persuade” experts to the opinion of the majority.
Scripting method. This is a comprehensive method for predicting complex processes with structural shifts. It consists in establishing a logically related sequence of events of a phased transition from the existing state of the forecasting object to the future state. Usually, when forecasting by the scenario method, there is always a time coordinate, i.e. the process unfolds in time. When writing scripts, they use the forward and reverse methods. With the direct method, the sequence of changes of events is built from the past to the future, i.e. the most likely successive changes in the forecasting object from its current state to the future are described. In the opposite way, first the future (desired) state of the forecasting object is described, then they try to formulate the nearest previous state from which it got there, and so consistently to the current state. A combination of the forward and reverse approaches usually yields the best results. Another of the secrets in forecasting scenarios is that it is necessary to start preparing scenarios with forecasting changes in the environment, and only then proceed to the description of the forecasting object. A qualitative scenario should reflect not only a consistent change in the object of forecasting, it should describe the problems and obstacles in achieving this state, the resources necessary to achieve the set goals, priorities in solving problems. In general, forecasting by the method of scenarios is largely an art and not a science, i.e. it is a difficult to formalize process, skills in the development of scenarios appear largely due to the experience and abilities of experts who develop forecasts.
Usually, forecasting by the scenario method is multivariate, most often two or three scenarios are made. With a two-variant, we are dealing with a pessimistic (the most unfavorable) scenario and an optimistic (most favorable) scenario. Between these options are all other options for the development of events. The expected value of the forecasted indicators is defined in this case as the arithmetic mean between the optimistic and pessimistic estimates.
More accurate results are usually given by a three-variant forecast, in this case, in addition to the two above scenarios, another one is developed – a probable scenario for the development of events. It means that this is a scenario in which all the helping and impeding factors will take their most common values. At first glance, it seems that the three-variant scenario is a two-variant yes plus the expected value found in the two-variant scenario. Far from it, in fact. The most probable scenario can and most often does not coincide with the average value of the optimistic and pessimistic options, in other words, the symmetry of the optimistic and pessimistic options relative to the probable is extremely rare. The expected value of the predicted value 
in the three-variant scenario of the development of events is determined by the following formula:
where: 
is the value of the predicted value in the pessimistic scenario of the development of events;
– the value of the predicted value in the most likely scenario;
– the value of the predicted value in the most favorable scenario.
Security questions
What are expert forecasting methods based on? Who is an expert? Describe the general scheme of expert forecasting. What is a qualitative assessment and in what cases is it applied? What methods of searching for experts do you know? What expert selection methods do you know which one is the best? What psychological difficulties arise in the commission method? Define the median. Define quartile. How to determine the consistency of expert opinion. Why is the median the best value for the final opinion of experts? Why is rank standardization necessary? What are the advantages and disadvantages of the questionnaire method? What are the advantages and disadvantages of the interview method? What are the advantages and disadvantages of the commission method? What are the advantages and disadvantages of the Delphi method? What is the effect of the first word? What is the essence of the Delphi method? What is a scenario and is it used for forecasting cases? What are optimistic, pessimistic, and probable options?