Development of adaptive combined models for predicting time series based on similarity identification
Journal Title: Восточно-Европейский журнал передовых технологий - Year 2018, Vol 1, Issue 4
Abstract
<p>Adaptive combined models of hybrid and selective types for prediction of time series on the basis of a program set of adaptive polynomial models of various orders were offered. Selection in these models is carried out according to B-, R-, P-criteria with automatic formation of the basic set of models based on the adaptive D-criterion. It was found that these models had the maximum accuracy in the case of short-term and medium-term prediction of time series.</p><p>Adaptive combined selective prediction models based on the R- and B-criteria of selection with identification of similarities in the retrospection of time series by the nearest neighbor method was proposed. An adaptive combined hybrid model of prediction with identification of similarities in the retrospection of time series was constructed. It was found that these models had the highest accuracy in the case of medium-term prediction of time series.</p><p>Estimation of the prediction efficiency of various combined models depending on the level of persistency of time series was made. It has been found that in the case of short-term prediction for the prediction period τ≤2, the adaptive combined hybrid prediction model is the most accurate. Selective models with various selection criteria are effective in predicting persistent time series with the Hurst index H>0.75 for the prediction period τ>2. In the case of prediction of time series with the Hurst index <img src="/public/site/images/anna1992/рис3.jpg" alt="" />for the prediction period τ>2, the adaptive combined hybrid and selective models with identification of similarities in the retrospection of the time series are more precise.</p>
Authors and Affiliations
Alexander Kuchansky, Andrii Biloshchytskyi, Yurii Andrashko, Svitlana Biloshchytska, Yevheniia Shabala, Oleksii Myronov
Keywords
Related Articles
- EP ID EP527751
- DOI 10.15587/1729-4061.2018.121620
- Views 71
- Downloads 0
Study of the method for assessing atmospheric turbulence by the envelope of sodar signals
<p class="a">The method for measuring the intensity of atmospheric turbulence based on assessment of statistical characteristics of the sodar signal envelope was investigated. This method requires no changes to the sodar...
The effect of using an energy accumulator on the level of emissions of pollutant substances by a shunting locomotive
Shunting locomotives at railroad stations account for most emissions of pollutants into the atmosphere. A significant share of these emissions depends on the position of the driver's controller that is used to operate a...
Study into the influence of operating processing parameters on qualitative characteristics of the carrot concentrated product
<span lang="EN-US">The process of concentrating vegetable juice in an improved vacuum evaporator with a stirrer which is simultaneously a heat exchanger was studied. A mathematical model describing kinetics of product he...
Investigation and analysis of the possibility of diffusionless phase transformations in the surface layer of a part under the action of grinding temperatures
<p>The possibility of diffusionless phase transformations in the surface layer of the ground part under the influence of instantaneous grinding temperature is investigated and analyzed. This is important because the phas...
Development of an algorithm for complex processing of geospatial data in the special-purpose geoinformation system in conditions of diversity and uncertainty of data
The algorithm of complex processing of geospatial data in special-purpose geoinformation systems in the conditions of diversity and uncertainty of data is developed. The novelty of the algorithm is to ensure the function...