Skip to main content
فهرست مقالات

modeling the impact of news on volatility: the case of iran

نویسنده:

ISC (20 صفحه - از 65 تا 84)

کلیدواژه ها :

news impact curve; Volatility; Stock Market

کلید واژه های ماشینی : ARCH، GARCH، In ARCH Iran Engle Ng، EGARCH، Iran، PNP، The EGARCH، In ARCH، Engle Ng، The PNP

In this paper various ARCH models and relevant news impact curves including a partially nonparametric (PNP) one are compared and estimated with daily Iran stock return data. Diagnostic tests imply the asymmetry of the volatility response to news. The EGARCH model, which passes all the tests and appears relatively matching with the asymmetry in the data, seems to be the most adequate characterization of the underlying data generating process. The PNP model successfully reveals the shape of the news impact curve and is a useful approach to modeling conditional heteroskedasticity.

خلاصه ماشینی: "Abstract In this paper various ARCH models and relevant news impact curves including a partially nonparametric (PNP) one are compared and estimated with daily Iran stock return data. In this paper we compare and estimate various ARCH models including a partially nonparametric one with daily Iran stock return data and use the news impact curve analysis of Engle and Ng (1993) in order to examine the relationship between return shocks and conditional volatility. Section V presents the partially non-parametric model of stock return volatility and estimates of the news impact curves. As a further diagnostic check for the adequacy of the various parameterizations of the conditional variance equations the moment type specification test suggested by Pagan and Sabau (1992) was computed from the regression (9) where and are the squared innovations and the estimated conditional variances, respectively, from the models reported in Table 3. The results of the ordinary least squares estimation of (9) presented in Table 4 suggest that just the EGARCH variance model pass moment specification test, Once again implying the GJR and GQARCH are a poor characterization of the underlying data generating process. Table 4: Moment Specification Test for the Estimated Conditional Variance Model GARCH EGARCH GJR GQARCH 0. The news impact curve estimates suggest that the GJR, GQARCH, and GARCH models are too extreme in the tails, and thus an inadequate characterization of the conditional variance of the Iran stock market."

  • دانلود HTML
  • دانلود PDF

برای مشاهده محتوای مقاله لازم است وارد پایگاه شوید. در صورتی که عضو نیستید از قسمت عضویت اقدام فرمایید.