Time series prediction based on averaging values via neural networks

Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61988987%3A17310%2F14%3AA1501BBA" target="_blank" >RIV/61988987:17310/14:A1501BBA - isvavai.cz</a>
Výsledek na webu
—
DOI - Digital Object Identifier
—

Jazyk výsledku
angličtina
Název v původním jazyce
Time series prediction based on averaging values via neural networks
Popis výsledku v původním jazyce
This paper introduces a development method for time series prediction based on averaging values. The experimental part will focus on teaching more neural networks with the same topology and settings that will solve the same problem (time series). The resulting values for training and test set are averaged depending on how many neural networks are involved in the calculation. The experimental part is focused on testing of periodic time series with different topologies and neural network settings. The results of prediction are compared with ARIMA models.
Název v anglickém jazyce
Time series prediction based on averaging values via neural networks
Popis výsledku anglicky
This paper introduces a development method for time series prediction based on averaging values. The experimental part will focus on teaching more neural networks with the same topology and settings that will solve the same problem (time series). The resulting values for training and test set are averaged depending on how many neural networks are involved in the calculation. The experimental part is focused on testing of periodic time series with different topologies and neural network settings. The results of prediction are compared with ARIMA models.

Rok uplatnění
2014
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Podobné výsledky(10)