A Signal Processing Model for Time Series Analysis: the Effects of International F/X Markets on Domestic Currencies Using Wavelet Networks

Cifter A., Ozun A.

INTERNATIONAL REVIEW OF ELECTRICAL ENGINEERING-IREE, cilt.3, sa.3, ss.580-591, 2008 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 3 Sayı: 3
  • Basım Tarihi: 2008
  • Dergi Adı: INTERNATIONAL REVIEW OF ELECTRICAL ENGINEERING-IREE
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.580-591
  • Anahtar Kelimeler: Signal processing, Wavelet networks, Neural networks, FX markets, PREDICTION
  • Marmara Üniversitesi Adresli: Evet

Özet

This paper proposes a powerful methodology wavelet networks to investigate the effects of international F/X markets on emerging markets currencies. We used EUR/USD parity as input indicator (international FIX markets) and three emerging markets currencies as output indicator (emerging markets currency). We test if the effects of international FIX markets change across different timescale. Using wavelet networks, it is found that the effects of international FIX markets increase with higher timescale. This evidence shows that the causality of international F/X markets on emerging markets should be tested based on 32-64 days effect. We also find that the effects of EUR/USD parity on Turkish Lira is higher on 9-16 days and 33-64 days scales and this evidence shows that Turkish lira is less stable compare to other emerging markets currencies as international F/X markets effects Turkish lira on shorten time scale. Besides it is found that Russian ruble is mostly affected from international FIX market in the long time according to wavelet networks analysis. Copyright (C) 2008 Praise Worthy Prize S.r.l. - All rights reserved.