The first statistical publication – the Fourth Book of Moses, “Numbers” in the Old Testament. We trace the development of ideas about the statistics until the twentieth century. The present stage of statistical methods began with parametric statistics by Pearson, Student, Fisher. Scientometrics of statistical researches provides an indication of the accumulated results. Nonparametric statistics appeared in the 1930s, applied statistics in our country - at the turn of 1970-80. We have discussed what gives applied statistics to national economy. Also we have told briefly about the history of statistical methods in
Russia (until Kolmogorov's time)
Some estimators of the probability density function in spaces of arbitrary nature are used for various tasks in statistics of non-numerical data. Systematic exposition of the theory of such estimators had a start in our work [2]. This article is a direct continuation of the article [2]. We will regularly use references to conditions and theorems of the article [2], in which we introduced several types of nonparametric estimators of the probability density. We studied more linear estimators. In this article we consider particular cases - kernel density estimates in spaces of arbitrary nature. When estimating the density of the one-dimensional random variable, kernel estimators become the Parzen-Rosenblatt estimators. Asymptotic behavior of kernel density estimators in the general case of an arbitrary nature spaces are devoted to Theorem 1 - 8. Under different conditions we prove the consistency and asymptotic normality of kernel density estimators. We have studied uniform convergence. We have introduced the concept of "preferred rate differences" and studied nuclear density estimators based on it. We have also introduced and studied natural affinity measures which are used in the analysis of the asymptotic behavior of kernel density estimators. We have found the asymptotic behavior of dispersions of kernel density estimators and considered the examples including kernel density estimators in finite-dimensional spaces and in the space of square-integrable functions
The article analyzes the development of the theory
of statistical control (from the XVIII century to the
present). Prof. M.V. Ostrogradskii (1846) clearly
describes the practical needs (ie, arising from the
quality assurance of large quantities of bags of
flour or pieces of cloth), to meet whom he spent his
research. At the same time Simpson was among the
ideas of probability theory XVIII century.
Therefore prof. M.V. Ostrogradskii may be
regarded as the founder of the theory of statistical
process control (not only in our country but all over
the world). Limit theorems of probability theory
and mathematical statistics have provided a
number of asymptotic results in problems of
statistical quality control, offer based on these best
practices. However, we must find out how much
interest among specialists characteristics are
different from limit for finite sample sizes. Such
research for the synthesis algorithm control plan on
the basis of the limit average output level of defects
is made in this article, and for the synthesis
algorithm control plan on the basis of the
acceptance and the rejection levels of defects - not
yet (clarification of the conditions of applicability
of this algorithm - unsolved problem of applied
mathematics). We have briefly reviewed the
development of our researches on the statistical
control. Control units can be not only some units of
production, but also documents (with internal and
external audit), and standard units of air, water and
soil in the environmental monitoring. One of the
achievements can be regarded as the transfer of
statistical control of production for environmental
monitoring
We have studied the asymptotic behavior of a broad class of nonparametric statistics, which includes statistics of omega-square type and Kolmogorov-Smirnov type. Limit theorems have been proved. We have also developed the method of approximation with step functions. With the help of this method we have obtained a number of necessary and sufficient conditions
The work discusses various examples of physical
systems which state is determined by the logarithmic
law - quantum and classical statistical systems and
relativistic motion in multidimensional spaces. It was
established that the Fermi-Dirac statistics and BoseEinstein-Maxwell-Boltzmann
distribution could be
described by a single equation, which follows from
Einstein's equations for systems with central
symmetry. We have built the rate of emergence of
classical and quantum systems. The interrelation
between statistical and dynamic parameters in
supergravity theory in spaces of arbitrary dimension
was established. It is shown that the description of the
motion of a large number of particles can be reduced
to the problem of motion on a hypersphere. Radial
motion in this model is reduced to the known
distributions of quantum and classical statistics. The
model of angular movement is reduced to a system of
nonlinear equations describing the interaction of a test
particle with sources logarithmic type. The HamiltonJacobi
equation was integrated under the most general
assumptions in the case of centrally-symmetric metric.
The dependence of actions on the system parameters
and metrics was found out. It is shown that in the case
of fermions the action reaches extremum in fourdimensional
space. In the case of bosons there is a
local extremum of action in spaces of any dimension
The article discusses various examples of dynamical
systems in which the motion is determined by the
logarithmic law - quark systems, hydrodynamic
systems, galaxies. Set the general nature of angular
motion on a hypersphere in a space of arbitrary
dimension and radial movement 6D in the metric of a
logarithmic potential. We investigate the 6D metric
describing the case of motion with two centers of
symmetry. It is shown that in such a metric exists a
class of exact solutions, logarithmically dependent on
the gravity center coordinates. It was established that
in spiral galaxies the orbital motion is due to the
logarithmic potential, which is the exact solution of the
field equations of Einstein's theory of gravity. The
most well-known and widespread in nature case is
turbulent flow over a smooth or rough surface, in
which the mean velocity depends logarithmically on
the distance from the wall. We derivate the logarithmic
velocity profile in turbulent flow from the NavierStokes
equations. An analogy of the logarithmic
velocity profile and the logarithmic law in the case of
erosion of materials under impacts been proposed. In
electrodynamics, Ampere's law, which describes the
interaction of current-carrying conductors, is a
consequence of the logarithmic dependence of the
vector potential of the distance from the conductor
axis. There is, however, an alternative derivation of
Ampere law of the Riemann hypothesis about the
currents due to the motion of charges
The question of extending the Lorentz electrodynamics to quantum theory is discussed. The system of equations of the Lorentz quantum electrodynamics was established
Chemical processes are often connected with use or
formation of condensed dispersed phase (CDP).
Dispersed particles can change mobility of charges, as
well as other parameters of the low-temperature plasma.
The aim of this work is to study the effect of magnetic
field on the processes of dispersed particles formation in
argon-oxygen plasma containing iron and carbon atoms
at atmospheric pressure. The equilibrium composition of
iron and carbon atoms containing mixture simulated at
temperatures of 1000-5000K for optimization of the
plasma-forming gas composition. It is shown that in case
of oxygen excess, the CDP particles contain only iron
oxides. The literature data about the phase transition
processes in a low-temperature plasma, as well as the
data about the processes with participation of
ferromagnetic particles in a constant magnetic field
analyzed. The results of investigations of the dispersed
particles forming in argon-oxygen plasma of arc
discharge in the presence and in the absence of the
magnetic field are shown. The formed disperse phase
was deposited on the substrates and studied by the
electron microscopy and X-ray methods. It was found
that with the lack of oxygen the size of the iron-oxide
particles created in the arc discharge containing iron and
carbon is affected by magnetic field: in a magnetic field
of 10 mT the particles are larger than in its absence
The new paradigm of mathematical statistics is based on the transition from parametric to nonparametric statistical methods, the numerical data - to non-numeric, on the intensive use of information technology. Its distinctive features are revealed in comparison with the old paradigm of mathematical statistics in the mid-twentieth century
This article focuses on the mathematical modeling of
evaluation of financial and economic activities of a
company and on definition (based on this model) of
such balance settings (line F1 and F2) which would
make financial-economic indicators of the activities of
the organization optimal, and the total cumulative
score was the maximum. The knowledge and the use
of the optimal parameters of the balance will allow the
managers to plan strategy for the future development
of the company. The article analyzes the dependencies
of each of the 15 basic indicators (profitability,
turnover, financial stability, liquidity and solvency) of
financial and economic activity of the organization on
the balance parameters. The optimal values of the
parameters of the balance and the main indicators of
financial and economic activities of the organization
have been found. We have also built a mathematical
model of optimal control of financial and economic
indicators in the form of a problem of mathematical
programming. For example, for the company called "Nika" it is shown the possibility of improving
estimation of financial and economic condition of the
organization. Knowledge of the optimal parameters of
the balance will allow the managers to plan strategy
for the future development of the organization. To
solve this problem we have used the method of
generalized reduced gradient implemented in Excel,
with which there was found a maximum of the
objective function for the article restrictions. The
article describes the analysis algorithm of the
optimization problem. A common assessment was
carried out in stages, based on the calculation
algorithm of sequentially improved target functions