ВОЗМОЖНОСТИ ИСПОЛЬЗОВАНИЯ МАРКОВСКОЙ МОДЕЛИ ДЛЯ СТАТИСТИЧЕСКОГО ОПИСАНИЯ ДИФФЕРЕНЦИАЛЬНОЙ РИТМОГРАММЫ
| Jazyk: | ruština |
|---|---|
| Rok vydání: | 2019 |
| Předmět: | |
| DOI: | 10.25686/2306-2819.2019.3.22 |
| Popis: | В статье рассматривается формирование марковской модели случайного процесса и оценки её параметров по поступающим реализациям. Описано применение марковских моделей для описания дифференциальной ритмограммы сердечнососудистой системы, выбраны интервалы квантования, приведены примеры. Получены оценки характеристик предлагаемой модели. Introduction. In various areas of human activities (medicine, technology, etc.), there are quasiperiodic processes. Their study and analysis require adequate statistical simulation models. The properties of theoretical models, for instance, the typical Gaussian random process, usually do not fit the processes in question. Thus, it is imperative to build models by the observed implementations of the processes. The aim of the research was to study the possibilities of application of Markov model for modeling differential rhythmogram. It seems rather promising to apply simple Markov model of a random process, which transition probability matrix contains a large number of parameters that can be estimated by the received process implementations. Conclusions. The model allows to display a variety of statistical and correlation properties. It is enabled to integrate estimates of the transition probabilities of the model by the onetype implementations of the random processes. Markov model can be employed in problems of classification of random processes by observed implementations. Analysis of the time properties (rhythmograms) of quasiperiodic processes and their modeling is conveniently carried out with the use of differences of adjoint periods (differential rhythmograms). №3(43) (2019) |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|