Kernel Ridge Regression with Odd Kernels in Financial Time Series Prediction

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F11%3A00181847" target="_blank" >RIV/68407700:21230/11:00181847 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Kernel Ridge Regression with Odd Kernels in Financial Time Series Prediction
Original language description
A new type of kernels called odd kernels is proposed to model dependencies by odd functions. Odd kernels could be simply constructed from the most of traditional kernels. Although usability of odd kernels is wider, we deals with their application to thefinancial time series prediction which was our primary motivation for construction of odd kernels. Applicability of odd kernels is demonstrated on high frequency time series from four real futures contracts. Kernel ridge regression with Gaussian kernel and its odd counterpart is applied to the prediction of short-time price differences. Results of conducted experiments show that frequently used Gaussian kernel can be outperformed by odd Gaussian kernel. Moreover, the results support the hypothesis of price differences symmetry.
Czech name
—
Czech description
—

Publication year
2011
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
POSTER 2011 - 15th International Student Conference on Electrical Engineering
ISBN
978-80-01-04806-1
ISSN
—
e-ISSN
—
Number of pages
1
Pages from-to
1
Publisher name
ČVUT, Fakulta elektrotechnická
Place of publication
Praha
Event location
Prague
Event date
May 12, 2011
Type of event by nationality
EUR - Evropská akce
UT code for WoS article
—

Similar results(10)