The problem of estimating the parameters of the model of a measurement experiment using the results of measurements with an error is considered. The mathematical model of the measurement error is formulated in terms of the theory of possibilities; the distribution of the possibilities on the set of error values determines the order that indicates which error values are preferred (that more probably occur during measurement) and which are less preferred. It is assumed that small error values are preferable to large ones. The mathematical model of the experiment depends on unknown parameters. The problem is to specify the values of these parameters by choosing their estimate for which the difference between the results of experiment and the model prediction is the most possible; this estimate is called the estimate of maximum possibility. An example of estimating the parameters of a Mössbauer spectrometric experiment is given.
07.05.Tp Computer modeling and simulation
02.70.-c Computational techniques; simulations
Department of Physics, Moscow State University, Moscow, 119991 Russia