In this paper we investigate volatility in the Nikkei daily index from the perspective of a modern view of time series analysis offered by wavelets. With wavelets and their multiresolution strength we can study the localization properties of the observed signal at both time and frequency domain, thus investigating the long and short term dynamics of the underlying volatility process. We keep our study at an exploratory level and try to understand, first of all, if wavelets can somehow help data to speak by themselves.
|Original language||English (US)|
|Number of pages||12|
|Journal||Neural Network World|
|State||Published - 1997|
ASJC Scopus subject areas
- Hardware and Architecture
- Artificial Intelligence