Effectiveness of Turkish Derivatives Market and Forecasting Comparative Prices for the Contracts

Taner TAŞ; Sibel SELİM

doi:https://doi.org/10.21121/eab.638575

Makale detayları

Effectiveness of Turkish Derivatives Market and Forecasting Comparative Prices for the Contracts

Yazarlar:

Taner TAŞ | Sibel SELİM

Yayıncı:

Ege University

Ana Başlıklar:

Ekonomi ve Maliye

Dergi:

Ege Academic Review 2019 Vol.19 Issue 4, pp.469-485

Belge Tipi:

Makale

Yayın Tarihi:

10/2019

Belge Dili:

İngilizce

Document ID (Org No.):

2000950673

Sıklık:

Üç Aylık

pISSN:

1303-099X

DOI:

https://doi.org/10.21121/eab.638575

Anahtar kelimeler:

Özet:

 


Derivative markets developed for eliminating uncertainty and risk arising from financial markets can make predictions about the future by using past price movements in case the market is not effective. In this context, in this study, firstly, the effectiveness of the Turkish Derivatives Market was tested by applying the Augmented Dickey-Fuller (ADF), Phillips-Perron (PP) and Kwiatkowski et al. (KPSS) linear unit root tests and Kapetanios et al. (KSS) nonlinear unit root test. As a result of all unit root tests, it was concluded that the series did not show random walk, so that the market was not effective. Then, the method that shows the highest performance is tried to be determined when forecasting the end of day settlement price of the TL/Dollar and Bist-30 contracts which is traded in the Derivatives Market. For this purpose, the forecasting results produced by the time series analysis methods are compared with the results of the artificial neural network model which has the best performance by employing different architectures, layer numbers, cell numbers in layers, activation functions and learning methods using the data which is provided from Borsa Istanbul Inc. and covering the dates between 04.02.2005 and 31.12.2015.According to the results of analysis, ARMA (4,4) model performed better than RBF-1-BL artificial neural network model and ARCH (1) model for TL/Dollar contract series. For the Bist-30 contract series, TDNN-1-B-L artificial neural network model has higher predictive performance than ARMA (4.5) and ARCH (1) models.

.

Makale detayları

Effectiveness of Turkish Derivatives Market and Forecasting Comparative Prices for the Contracts

Yazarlar:

Taner TAŞ | Sibel SELİM

Yayıncı:

Ege University

Ana Başlıklar:

Ekonomi ve Maliye

Dergi:

Ege Academic Review 2019 Vol.19 Issue 4, pp.469-485

Belge Tipi:

Makale

Yayın Tarihi:

10/2019

Belge Dili:

İngilizce

Document ID (Org No.):

2000950673

Sıklık:

Üç Aylık

pISSN:

1303-099X

DOI:

https://doi.org/10.21121/eab.638575

Türev Piyasalar

Etkin Piyasalar Hipotezi

Birim Kök Testi

ARIMA Modelleri

ARCH/ GARCH Modelleri

Yapay Sinir Ağları

Sayı

Makale detayları

Effectiveness of Turkish Derivatives Market and Forecasting Comparative Prices for the Contracts

Yazarlar: Taner TAŞ | Sibel SELİM

Yayıncı: Ege University

Ana Başlıklar: Ekonomi ve Maliye

Dergi: Ege Academic Review 2019 Vol.19 Issue 4, pp.469-485

Belge Tipi: Makale

Yayın Tarihi: 10/2019

Belge Dili: İngilizce

Document ID (Org No.): 2000950673

Sıklık: Üç Aylık

pISSN: 1303-099X

DOI: https://doi.org/10.21121/eab.638575

Türev Piyasalar

Etkin Piyasalar Hipotezi

Birim Kök Testi

ARIMA Modelleri

ARCH/ GARCH Modelleri

Yapay Sinir Ağları

Sayı

Yazarlar:

Taner TAŞ | Sibel SELİM

Yayıncı:

Ege University

Ana Başlıklar:

Ekonomi ve Maliye

Dergi:

Ege Academic Review 2019 Vol.19 Issue 4, pp.469-485

Belge Tipi:

Makale

Yayın Tarihi:

10/2019

Belge Dili:

İngilizce

Document ID (Org No.):

2000950673

Sıklık:

Üç Aylık

pISSN:

1303-099X

DOI:

https://doi.org/10.21121/eab.638575