Authors

Imitation modeling at the choice of composition of factors and structure

It was demonstrated that at the choice of the best of the built variants of regression model the orientation is only to the value of coefficient of determination and Fcriterion can’t warrant obtaining of minimal error of forecast of response function. For comparative assessment of quality of obtained variants of regression equation was demonstrated the opportunity of usage of developed program systems which allow at minimum labour costs to build imitation models, to carry out imitation modeling, to estimate statistic characteristics of response function (expectation value, dispersion, coefficient of variation of excess, unbalance and character of distribution) at various values of errors of covariates and different structure of regression model. Knowing the character of allocation of response function, we can estimate the probability of that the response function will be found within the prescribed limits. As an example related to the solution of real applied problem it is demonstrated that depending on the structure of regression equation and accuracy of measurement or fixation of covariates the value of response function error can vary over a wide range.

Evaluating the reserves to reduce resource consumption for producing goods and services: methods and tools

A method for estimating reserves reducing the resource consumption of goods and services which requires minimal resources is suggested. It is shown that the simulation may provide obtaining both statistical characteristics and the overhead distribution of each resource-allocation process. It is proposed to estimate reserves to reduce resource consumption of goods and services on the magnitude of the coefficient of variation, the value of right-sided asymmetry and the median of the empirical distribution of the resource costs.