Forecasting using probabilistic models is based on the exponential smoothing method. Probabilistic models are inherently different from extrapolation time series models, in which the basis is the description of a change in process time. In time series, the models represent some
function of time with coefficients, the values of which are estimated from observations. In probabilistic models, probabilities are estimated, not coefficients.
The interview method involves a conversation between a forecaster and an expert, during which the forecaster in accordance with a previously developed the program poses questions to the expert regarding the prospects for the development of the predicted object. The success of such an assessment largely depends on the ability of the interviewed expert to give impromptu conclusions on a variety of fundamental issues.
Analytical expert evaluations of a long and thorough independent work of an expert on analysis of trends, assessment of the state and development paths of the predicted object. This method enables the expert to use special information about the forecast object. Their the expert draws up the considerations in the form of a memorandum.
Comments
Leave a comment