Non-parametric sign test and paired samples test of effectiveness of official FX intervention
| Autor: | Srdjan Marinkovic |
|---|---|
| Rok vydání: | 2014 |
| Předmět: |
Actuarial science
Foreign exchange market sterilized FX interventions exchange rate dynamics NBS event study methodology Nonparametric statistics Event study jel:F31 sterilized FX interventions jel:E58 National bank lcsh:HD72-88 exchange rate dynamics lcsh:Economic growth development planning Test (assessment) NBS Exchange rate jel:G15 foreign exchange market Economics Econometrics Sign test event study methodology General Economics Econometrics and Finance Foreign exchange market Event (probability theory) |
| Zdroj: | Ekonomski Anali, Vol 59, Iss 202, Pp 107-130 (2014) |
| ISSN: | 1820-7375 0013-3264 |
| DOI: | 10.2298/eka1402107m |
| Popis: | This article is an event study of the effectiveness of official foreign exchange interventions by the National Bank of Serbia (NBS) in the RSD/EUR market. As the NBS does not have a formally modelled response function we assume that it intervenes as is expected according to its mandate, i.e., to prevent excess daily fluctuations. This paper tests two alternative goals of official intervention, marking as a success an event in which the NBS either breaks/reverses or smooths an ongoing exchange rate movement. According to a pre-defined time window for an intervention-clustering event, a nonparametric sign test supported the view that the NBS has failed to reverse the trend but is fairly effective in smoothing exchange rate return. However, even the smoothing effect is identified as short lasting. A paired samples test leads to similar findings, but because of weak support for the necessary conditions of sampling distribution it remains less conclusive. [Projekat Ministarstva nauke Republike Srbije, br. 179015: Challenges and prospects of structural changes in Serbia: Strategic directions for economic development and harmonization with EU requirements] |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|