Abstract

The possibilities of an improved autoregression model and an integrated moving average (ARMAS) for the analysis of non-stationary data and the identification of long-term trends in the processes under study are considered. The proposed model can be used to study the observed processes in various areas of human activity: the analysis of the observed trajectories of the movement of aircraft, in particular unmanned aerial vehicles, meteorological processes that reflect the state of the atmosphere. The mathematical apparatus developed in the article was used to analyze changes in the atmospheric temperature time series observed for a long time, the average annual temperatures were estimated, followed by sliding smoothing with a low-frequency filter.
 It is shown that the removal of the seasonal component in the ARPSS model eliminates or distorts significantly the trend and has little effect on the stationary component of the ARPSS process. The operation of de-trending has little effect on the properties of the seasonal component and the stationary component of the process. To assess the trend, the mean annual temperatures were preliminarily obtained. The use of moving averaging, which removes the seasonal component from the average monthly temperatures, makes it possible to find a weak long-term trend. The results obtained in the work can be used to analyze medium-term and long-term changes in atmospheric phenomena, to refine the results obtained by traditional methods of processing results and methods of mathematical statistics, as well as in other areas of human activity.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call