Search

The Julia programming language is a specialized language for scientific computing. It is relatively new, so most of the libraries for it are in the active development stage. In this article, the authors consider the possibilities of the language in the field of mathematical statistics. Special emphasis is placed on the technical component, in particular, the process of installing and configuring the software environment is described in detail. Since users of the Julia language are often not professional programmers, technical issues in setting up the software environment can cause difficulties that prevent them from quickly mastering the basic features of the language. The article also describes some features of Julia that distinguish it from other popular languages used for scientific computing. The third part of the article provides an overview of the two main libraries for mathematical statistics. The emphasis is again on the technical side in order to give the reader an idea of the general possibilities of the language in the field of mathematical statistics.

Regression equations are derived for the calculation of roll springing of ARWS-1, ARWS-1a, ARWS-2 and ARWS-2a forging rolls on the ground of experimental results referring to the production conditions effect and statistical data processing. They are included in a design technique focused on impressions of sector-dies for two-, three- and four passes hot forge-rolling. [ABSTRACT FROM AUTHOR]

A method for the statistical processing of the input data on the stress state of the Earth's lithosphere that takes the initial 3-D position of the principle stress axes into account is elaborated. This approach is based on the calculation of the arithmetic mean value for every six independent tensor components during determination of the average stress in any sampling. When determining the sampling for calculation of the average stress for the current cell of the calculated grid, it is proposed to insert the measurements into the sampling that are spatially located in such a manner where the distance from the measurement point to the cell center is less than some value that is named as a search radius. The latter either was specified as a constant for all cells of the calculated grid or was determined assuming that the dispersion of the average tensor was less than some preset value. The results of the application of this approach are presented based on the example of the processing of the measurements from the World Stress Map (Heidbach et al., 2008). The resultant set of the mean stress-field maps reflects the generalized pattern of the stress distribution in the Earth's lithosphere. [ABSTRACT FROM AUTHOR]