Similarity of Stock Market Series Analyzed in Hilbert Space

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21340%2F16%3A00305204" target="_blank" >RIV/68407700:21340/16:00305204 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Similarity of Stock Market Series Analyzed in Hilbert Space
Original language description
Time series from stock markets exhibit various types of fluctuations and can be studied as samples from unknown n-dimensional distribution. The main question behind the paper is how to recognize similar stocks or similar time periods of their history. Whenever current stock segment is similar to another former one of the same stock or the other, the investment strategy should be also similar. Parzen estimate of probability density function was used to obtain scalar products of stock segment pairs. This product and corresponding metrics were expressed analytically as double sum over segment items to enable the use of Principal Component Analysis, Cluster Analysis and Self-Organized Maps for the description of stock similarities and differences. Mathematical formulation of similarity task and analytic calculations are results of my original research work. General principle is demonstrated on real stocks of leading IT companies in the period 2009-2015.
Czech name
—
Czech description
—

Project
—
Continuities
S - Specificky vyzkum na vysokych skolach<br>I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Publication year
2016
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Similar results(10)