Seasonal Long-Term Dependence: Evidences from the Bangladeshi Financial Series | ||||
| The Egyptian Statistical Journal | ||||
| Article 8, Volume 51, Issue 2, December 2007, Page 135-146 PDF (2.39 MB) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/esju.2007.313844 | ||||
 View on SCiNiTO | ||||
| Author | ||||
| Shipra Banik* | ||||
| School of Engineering & Computer Science, Independent University, Bangladesh, Dhaka | ||||
| Abstract | ||||
| It is well-known that many time series exhibit strong-dependence long memory property. In this paper, the seasonal structure of the Bangladeshi quarterly financial series is examined by means of recently proposed fractional integration test. Evidences suggest that most of the series are seasonal long-term dependent. Thus, the standard approach of taking seasonal differences to get the series stationary at different frequencies (which is required) may lead to spurious results, because outcomes suggest that the series having a component of long memory behavior. Findings of this study are useful to the following groups: (i) Policy makers who are interested to make wise financial policies, (ii) different practitioners whose success depends on the ability to predict financial series and (iii) applied researchers who want to improve the model specifications of the selected series. | ||||
| Keywords | ||||
| Long-Memory - Quarterly Unit Roots; Seasonality - Financial Series | ||||
|
Statistics Article View: 105 PDF Download: 17 |
||||
