Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/238792 
Year of Publication: 
2010
Citation: 
[Journal:] International Econometric Review (IER) [ISSN:] 1308-8815 [Volume:] 2 [Issue:] 2 [Publisher:] Econometric Research Association (ERA) [Place:] Ankara [Year:] 2010 [Pages:] 73-96
Publisher: 
Econometric Research Association (ERA), Ankara
Abstract: 
We study time-varying realized volatility and related correlation measures as proxies for the true volatility and correlation. We investigate measures of Two-Scale realized Absolute Volatility (TSAV) and correlation (TSACORxy) which are helpful to cope effectively with the problem of market microstructure effects at very high frequency financial time series. The measures are constructed based on subsampling and averaging method so that they possess rather less bias even in presence of market microstructure noise. Absolute transformation of return values has been proved in literature to be more robust than squared transformation when considering large values. With respect to some stylized facts of markets, realized squared correlation does not display dynamic behavior. Motivated by robustness of realized absolute volatility, we study an alternative measure of correlation, built on absolute-transformed volatility. This measure of correlation exhibits experimentally some dynamics and hence some predictability capability on minute-by-minute frequency exchange market data. We show that the distribution of realized correlation series computed based on TSACORxy tends to comply a rightward asymmetric shape implying that upside co-movements are greater than downside ones. Moreover we study the association between realized volatility and correlation. According to the two-scale measure, our findings empirically suggest that when returns in Euro/USD exchange rate are highly volatile, the relation between Euro/USD and Euro/GBP exchange markets is strong, and when Euro/USD calms down, the relationship relaxes.
Subjects: 
Realized Volatility and Correlation
Long Memory
Scaling Law
Self-Similarity Dimension
Market Microstructure Effects
JEL: 
C14
C51
C58
F31
G15
Creative Commons License: 
cc-by-nc-nd Logo
Document Type: 
Article

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.