Long Memory in Stock Returns and Volatility in India : A Nonparametric Analysis

By: Material type: ArticleArticleLanguage: ENG Series: ; 14Publication details: Dec 2008 0Edition: 12Description: 34-53 PpSubject(s): DDC classification:
  •  Bad
Online resources: Summary: This paper tests the long memory in daily aggregate returns on S&P CNX Nifty index and its volatility using Rescaled Range (R/S)-type nonparametric tests. The results do not show long memory in returns, but their squared and absolute values (which represent the volatility) show robust presence of long-range dependence in the data for the entire sample period. However, in a subsample covering the period from March 2001 to December 2007, the volatility measures do not show long memory. Since the tests of long memory cannot differentiate between fractional integration and structural breaks, absence of long memory in a subsample appears more consistent with the hypothesis of structural breaks in volatility dynamics.
Tags from this library: No tags from this library for this title. Log in to add tags.
Star ratings
    Average rating: 0.0 (0 votes)

This paper tests the long memory in daily aggregate returns on S&P CNX Nifty index and its volatility using Rescaled Range (R/S)-type nonparametric tests. The results do not show long memory in returns, but their squared and absolute values (which represent the volatility) show robust presence of long-range dependence in the data for the entire sample period. However, in a subsample covering the period from March 2001 to December 2007, the volatility measures do not show long memory. Since the tests of long memory cannot differentiate between fractional integration and structural breaks, absence of long memory in a subsample appears more consistent with the hypothesis of structural breaks in volatility dynamics.

There are no comments on this title.

to post a comment.

Powered by Koha