High Frequency Data: Making Forecasts and Looking for an Optimal Forecasting Horizon
Result description
We illustrate the AutoRegessiveiGeneralised collditionaliy Heteros§cedastic (ARCH-GARCH) methodology on the developing a forecast mode! for exchange rates time series of the Czech crown (CZK) againsr the Slovak crown (SKK) and make comparisons the forecast accuracy with the class of Radial Basic Functiov Neural neural network RBF NN models. To illustrate the forecasting performance of these approaches the inIJu t/r.utput function estimation based on REF networks is presented. III a comparative study isshown that the RBF NN approach is able to model and predkt high frequency data with reasol1abie accuracy and more efficient than statistical methods. In order to find the optima! forecasting horizon, we use the analysis of fo recast errors and choose tntvalues that give the smallest error variance. It is found that the error variance estimation process based on soft methods is simplified and less critical to the question whether the data is true crisp or white noise.
Keywords
time seriesARCH-GARCH modelssoft neural networksgranular computingforecast accuracy
The result's identifiers
Result code in IS VaVaI
Alternative codes found
RIV/47813059:19240/10:#0003230 RIV/63468352:_____/10:#0000301
Result on the web
—
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
High Frequency Data: Making Forecasts and Looking for an Optimal Forecasting Horizon
Original language description
We illustrate the AutoRegessiveiGeneralised collditionaliy Heteros§cedastic (ARCH-GARCH) methodology on the developing a forecast mode! for exchange rates time series of the Czech crown (CZK) againsr the Slovak crown (SKK) and make comparisons the forecast accuracy with the class of Radial Basic Functiov Neural neural network RBF NN models. To illustrate the forecasting performance of these approaches the inIJu t/r.utput function estimation based on REF networks is presented. III a comparative study isshown that the RBF NN approach is able to model and predkt high frequency data with reasol1abie accuracy and more efficient than statistical methods. In order to find the optima! forecasting horizon, we use the analysis of fo recast errors and choose tntvalues that give the smallest error variance. It is found that the error variance estimation process based on soft methods is simplified and less critical to the question whether the data is true crisp or white noise.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
IN - Informatics
OECD FORD branch
—
Result continuities
Project
Continuities
N - Vyzkumna aktivita podporovana z neverejnych zdroju
Others
Publication year
2010
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Sixth International Conference on Natural Computation
ISBN
978-1-4244-5960-5
ISSN
—
e-ISSN
—
Number of pages
6
Pages from-to
—
Publisher name
—
Place of publication
China
Event location
Yantai
Event date
Jan 1, 2010
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—
Result type
D - Article in proceedings
CEP
IN - Informatics
Year of implementation
2010