Detecting Major Economic Events from Short Financial Return Series Using Singular Spectrum Analysis
Research Mentor(s)
Noguchi, Kimihiro
Description
Although the majority of financial time series analysis focuses on the volatility (e.g., standard deviation) of returns, several recent studies focus on the return series themselves to understand their behaviors. Daily return series which consist of at least one thousand observations tend to contain useful information about short- and long-term cyclical patterns, reflecting weekend, intramonth, quarterly, and annual effects. In this study, we focus on short return series with less than three hundred observations to examine cyclical patterns as well as short-lived trend patterns. Using singular spectrum analysis (SSA), we suggest that the structure of short return series is rather complex, possibly consisting of more than twenty significant cyclical and trend components. Moreover, by examining swaps in the relative significance of these components, we investigate the relationship between major economic events and periods at which these swaps occur frequently. The results suggest that an appropriately weighted mean squared difference caused by possible swaps in the components well correlates with major economic events, supporting the view of interpreting volatility as a consequence of underlying structural changes in the trend or cyclical components of return series.
Document Type
Event
Start Date
18-5-2020 12:00 AM
End Date
22-5-2020 12:00 AM
Department
Statistics, Mathematics
Genre/Form
student projects, posters
Subjects – Topical (LCSH)
Economics--Statistics; Management science
Type
Image
Rights
Copying of this document in whole or in part is allowable only for scholarly purposes. It is understood, however, that any copying or publication of this document for commercial purposes, or for financial gain, shall not be allowed without the author’s written permission.
Language
English
Format
application/pdf
Detecting Major Economic Events from Short Financial Return Series Using Singular Spectrum Analysis
Although the majority of financial time series analysis focuses on the volatility (e.g., standard deviation) of returns, several recent studies focus on the return series themselves to understand their behaviors. Daily return series which consist of at least one thousand observations tend to contain useful information about short- and long-term cyclical patterns, reflecting weekend, intramonth, quarterly, and annual effects. In this study, we focus on short return series with less than three hundred observations to examine cyclical patterns as well as short-lived trend patterns. Using singular spectrum analysis (SSA), we suggest that the structure of short return series is rather complex, possibly consisting of more than twenty significant cyclical and trend components. Moreover, by examining swaps in the relative significance of these components, we investigate the relationship between major economic events and periods at which these swaps occur frequently. The results suggest that an appropriately weighted mean squared difference caused by possible swaps in the components well correlates with major economic events, supporting the view of interpreting volatility as a consequence of underlying structural changes in the trend or cyclical components of return series.