We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
USE OF NEURAL NETWORKS FOR PREDICTING DEVELOPMENT OF USA EXPORT TO CHINA TAKING INTO ACCOUNT TIME SERIES SEASONALITY.
- Authors
ROUSEK, PAVEL; MAREČEK, JAN
- Abstract
The objective of the contribution is to propose a methodology of taking into consideration the seasonal fluctuations in time series equalization using artificial neural networks on the example of the United States of America export to the People´s Republic of China. For the research, the data from the period between January 1985 and August 2018 are used. For the prediction, two types of neural networks and two variants of input data sets are used. In the second variant, the seasonal fluctuation is represented by a categorical variable. It resulted that all retained structures are applicable, but the retained MLP networks of the B alternative achieve better results. It has been proven that with the use of artificial neural networks, it is possible to predict the export development efficiency and with a high degree of accuracy, especially in the short term and considering specific seasonal fluctuations.
- Subjects
CHINA; UNITED States; ARTIFICIAL neural networks; EXPORTS; RADIAL basis functions
- Publication
Ad Alta: Journal of Interdisciplinary Research, 2019, Vol 9, Issue 2, p299
- ISSN
1804-7890
- Publication type
Article