Determination of Oil Rents in Turkey, India and Brazil with Autoregressive Fractionally Integrated Moving Average (ARFIMA) Model

Semanur Sarıçam; Barış Aşıkgil

Makale detayları

Determination of Oil Rents in Turkey, India and Brazil with Autoregressive Fractionally Integrated Moving Average (ARFIMA) Model

Yazarlar:

Semanur Sarıçam | Barış Aşıkgil

Yayıncı:

Marmara University

Ana Başlıklar:

Uygulamalı İstatistik

Dergi:

International Journal of Advances in Engineering and Pure Sciences Vol.37 Issue UYIK 2024 Special Issue, pp.21-29

Belge Tipi:

Makale

Yayın Tarihi:

1/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001331368

pISSN:

2636-8277

Anahtar kelimeler:

Özet:

 Oil rents are considered as part of energy policies as they are revenues from taxes on production, imports and consumption. The energy policies of countries such as Turkey, India and Brazil are part of the revenue generation methods of these countries due to high taxes on petroleum products. Considering the important role of oil rents in the economy, this study aims to analyse the oil rent data of these countries covering the years 1970-2016 with the autoregressive fractionally integrated moving average ARFIMA model, which is one of the long memory models. In the first stage, classical unit root tests are applied to determine whether the oil rent series of Turkey, India and Brazil are stationary. Then, Hursts RS statistic is used to assess whether the series have long memory. In the second stage, all possible autoregressive moving average ARIMA and ARFIMA models are tested for p and q values less than or equal to 2 In order to determine the best model, two basic information criteria Akaike Information Criterion AIC and Bayesian Information Criterion BIC are used. After the most appropriate model is determined, forecasts are made by using this model.

 Oil rents are considered as part of energy policies as they are revenues from taxes on production, imports and consumption. The energy policies of countries such as Turkey, India and Brazil are part of the revenue generation methods of these countries due to high taxes on petroleum products. Considering the important role of oil rents in the economy, this study aims to analyse the oil rent data of these countries covering the years 1970-2016 with the autoregressive fractionally integrated moving average ARFIMA model, which is one of the long memory models. In the first stage, classical unit root tests are applied to determine whether the oil rent series of Turkey, India and Brazil are stationary. Then, Hursts RS statistic is used to assess whether the series have long memory. In the second stage, all possible autoregressive moving average ARIMA and ARFIMA models are tested for p and q values less than or equal to 2 In order to determine the best model, two basic information criteria Akaike Information Criterion AIC and Bayesian Information Criterion BIC are used. After the most appropriate model is determined, forecasts are made by using this model.

 Oil rents are considered as part of energy policies as they are revenues from taxes on production, imports and consumption. The energy policies of countries such as Turkey, India and Brazil are part of the revenue generation methods of these countries due to high taxes on petroleum products. Considering the important role of oil rents in the economy, this study aims to analyse the oil rent data of these countries covering the years 1970-2016 with the autoregressive fractionally integrated moving average ARFIMA model, which is one of the long memory models. In the first stage, classical unit root tests are applied to determine whether the oil rent series of Turkey, India and Brazil are stationary. Then, Hursts RS statistic is used to assess whether the series have long memory. In the second stage, all possible autoregressive moving average ARIMA and ARFIMA models are tested for p and q values less than or equal to 2 In order to determine the best model, two basic information criteria Akaike Information Criterion AIC and Bayesian Information Criterion BIC are used. After the most appropriate model is determined, forecasts are made by using this model.

 Petrol kiraları, üretim, ithalat ve tüketim vergileri üzerinden elde edilen gelir olduğu için enerji politikalarının bir parçası olarak değerlendirilir. Türkiye, Hindistan ve Brezilya gibi ülkelerin enerji politikaları, petrol ürünlerine yüksek vergiler uygulanmasından dolayı bu ülkelerin gelir elde etme yöntemlerinin bir parçasıdır. Petrol kiralarının ekonomide önemli rolü göz önüne alındığında, bu çalışma, bu ülkelerin 1970-2016 yıllarını kapsayan petrol kira verilerini kullanılarak uzun hafıza modellerinden biri olan otoregresif kesirli bütünleşik hareketli ortalama ARFIMA modeliyle incelemeyi amaçlamaktır. İlk aşamada, Türkiye, Hindistan ve Brezilyanın petrol kira serilerinin durağan olup olmadığını belirlemek için klasik birim kök testleri uygulanmıştır. Sonrasında, serilerin uzun hafızaya sahip olup olmadığını değerlendirmek amacıyla Hurstün RS istatistiği kullanılmıştır. İkinci aşamada ise p ve q değerleri 2den küçük veya eşit olacak şekilde olası tüm otoregresif hareketli bütünleşik ortalama ARIMA ve ARFIMA modelleri test edilmiştir. En iyi model belirlenebilmesi için Akaike Bilgi Kriteri AIC ve Bayesci Bilgi Kriteri BIC iki temel bilgi kriterlerinden yararlanılmıştır. En uygun model belirlendikten sonra, bu model aracılığıyla geleceğe yönelik tahminler yapılmıştır.

Makale detayları

Determination of Oil Rents in Turkey, India and Brazil with Autoregressive Fractionally Integrated Moving Average (ARFIMA) Model

Yazarlar:

Semanur Sarıçam | Barış Aşıkgil

Yayıncı:

Marmara University

Ana Başlıklar:

Uygulamalı İstatistik

Dergi:

International Journal of Advances in Engineering and Pure Sciences Vol.37 Issue UYIK 2024 Special Issue, pp.21-29

Belge Tipi:

Makale

Yayın Tarihi:

1/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001331368

pISSN:

2636-8277

Oil rents

Long memory models

ARFIMA model

ARIMA model.Petrol kiraları

Uzun hafıza modelleri

ARFIMA modeli

ARIMA modeli.

Sayı

Makale detayları

Yazarlar: Semanur Sarıçam | Barış Aşıkgil

Yayıncı: Marmara University

Ana Başlıklar: Uygulamalı İstatistik

Dergi: International Journal of Advances in Engineering and Pure Sciences Vol.37 Issue UYIK 2024 Special Issue, pp.21-29

Belge Tipi: Makale

Yayın Tarihi: 1/2025

Belge Dili: Türkçe

Document ID (Org No.): 2001331368

pISSN: 2636-8277

Sayı

Yazarlar:

Semanur Sarıçam | Barış Aşıkgil

Yayıncı:

Marmara University

Ana Başlıklar:

Uygulamalı İstatistik

Dergi:

International Journal of Advances in Engineering and Pure Sciences Vol.37 Issue UYIK 2024 Special Issue, pp.21-29

Belge Tipi:

Makale

Yayın Tarihi:

1/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001331368

pISSN:

2636-8277