# NONLINEAR VARIANTS OF STATISTICAL DATA ANALYSIS

Considers possible representations of statistical data using mathematical formulas that characterize the results of scientific research / observations, related, in essence, to different physical or chemical quantities. At the same time, their operating time is matched either with the deterioration of parts with the passage of the product's operating time or with the change / decrease in the mass of the parts in a function of either time or friction path. The article draws attention to the formation and analysis of the bank of statistical data accumulated by the researcher when observing any physical phenomenon, say, for the volume of emissions of gases from enterprises and from the exhaust pipes of cars, the wear and tear of parts in the operation of machinery and machinery, pressure, temperature, velocity, etc. In this case, cases that correspond to a nonlinear dependence are considered, for an even and odd number of observation points for the operation of an object, system or pairs of friction as time is working or as path function. Computer calculations using the mathematical shell of MathCad are shown. In addition, an algorithm for finding the equation determining the best variant of approximating the theoretical approximation of the initial data to statistical information as quickly as possible is explained. In addition, a possible variant is considered when an uncharacteristic value occurs in statistical data. In this case, a special criterion is described in detail, which allows more accurate calculations [1]–[4].

**Authors:** K. N. Voinov, T. Nasr, A. Hildayati, Ya. Ghellab

**Direction:** Metrology and Metrologycal Assurance

**Keywords:** Statistical data, approximation, data bank, computer processing, information

View full article