J 2017

Comparison of neural networks and regression time series in estimating the development of the afternoon price of palladium on the New York Stock Exchange

VOCHOZKA, Marek

Basic information

Original name

Comparison of neural networks and regression time series in estimating the development of the afternoon price of palladium on the New York Stock Exchange

Name in Czech

Komparace neuronových sítí a regresních časových řad při odhadu vývoje odpoledních cen palladia na Newyorské burze

Authors

Edition

Trends Economics and Management, Brno, Fakulta podnikatelská Vysokého učení technického v Brně, 2017, 1802-8527

Other information

Language

English

Type of outcome

Article in a journal

Field of Study

50200 5.2 Economics and Business

Country of publisher

Czech Republic

Confidentiality degree

is not subject to a state or trade secret

Organization unit

Institute of Technology and Business in České Budějovice

Keywords (in Czech)

palladium; umělé neuronové sítě; regresní časové řady; predikce; vícevrstvé perceptronové sítě

Keywords in English

palladium; artificial neural networks; regression time series; prediction; multilayer perceptron network

Tags

Changed: 18/4/2018 08:18, Mgr. Blanka Mikšíková

Abstract

In the original language

The aim of the paper is to perform a regression analysis of the development of afternoon platinum prices on the New York Stock Exchange using neural networks and a simple linear regression. The partial aim is to compare these two methods and determine the most suitable ones for predicting the future development of afternoon platinum prices on the New York Stock Exchange. The analysis is made on the data on afternoon platinum prices in a time period exceeding 10 years. The result is a prediction of afternoon platinum prices and the fact that neural networks are more suitable than simple linear regression for this prediction.

In Czech

The aim of the paper is to perform a regression analysis of the development of afternoon platinum prices on the New York Stock Exchange using neural networks and a simple linear regression. The partial aim is to compare these two methods and determine the most suitable ones for predicting the future development of afternoon platinum prices on the New York Stock Exchange. The analysis is made on the data on afternoon platinum prices in a time period exceeding 10 years. The result is a prediction of afternoon platinum prices and the fact that neural networks are more suitable than simple linear regression for this prediction.

Files attached

Vochozka_-_comparison_of_neural_networks....pdf
Request the author's version of the file