Ülke kredi derecelendirmesine ilişkin farklı yöntem denemeleri

Abstract

Piyasaların ihtiyacı üzerine ortaya çıkan kredi derecelendirme, zamanla finansal piyasaların karmaşıklığı ve asimetrik bilgi artışı ile önemi daha da artmıştır. Derecelendirme çeşitlerinden ülke kredi derecelendirmesi, KDK'lar (Kredi Derecelendirme Kuruluşları) tarafından ülkelerin ekonomik ve politik risklerini göz önünde bulundurarak geliştirdikleri nicel ve nitel özellikleri puanlama yöntemleri ile belirlenmektedir. Yatırımcı kararlarına etkisi ile ülkeleri ekonomik açıdan etkileyen derecelendirme, birçok tartışmaya konu olmuştur. Bu çalışmada pazar payının çoğunluğunu oluşturan ve uluslararası kabul görmüş olan Standard and Poor's, Moody's ve Fitch'in ülke kredi derecelendirme metodolojileri incelenmiştir. Literatürdeki kredi derecelendirme konulu çalışmalar analiz edilmiştir. Günümüzde algoritmik ticaret ve yüksek frekanslı alım-satım işlemlerinin artması ile kısa vadede ülke ve şirket ekonomilerinin öngörülebilirliğinin önemi artmıştır. Daha hızlı ve kolay hesaplanabilenen, şeffaf, objektif ve sürekli güncellenebilen kredi derecelendirme modellerine ihtiyaç olduğu gözlemlenmiştir. Bu çalışmanın amacı ülkeleri kısa vadede yatırım yapılabilir ve yatırım yapılamaz olarak sınıflandıran iki model önermektir. İlk model, borsa günlük kapanış değerlerinden faydalanarak YSA (Yapay Sinir Ağları) yöntemi ile geçerli kredi derecelerine göre eğitilerek ülkeleri sınıflandırmaktadır. Bu model ile elde ettiğimiz sonuçlar Standard and Poor's geçerli dönem kredi dereceleri ile tamamen örtüşmektedir. Fitch ile karşılaştırıldığında ise Rusya dışında YSA sınıflandırması ile uyum sağlamaktadır. Diğer model ise iki aşamadan oluşmaktadır. İlk aşamada aylık makroekonomik değişkenler ve geçmiş borsa endeks değerleri ile NLMS (Normalized Least Mean Square - Normalize En Küçük Ortalama Kareler) yöntemi kullanılarak borsa değeri tahmini yapılmaktadır. İkinci aşamada tahmin sonuçlarına göre ülkelerin sınıflandırma işlemi gerçekleşmektedir. Bu modelden elde ettiğimiz sonuçlar Fitch ile tamamen örtüşmektedir, Standard and Poor's ile karşılaştırdığımızda ise Türkiye ve Portekiz için farklı sonuçlar elde edilmiştir.Bu çalışmada standart yaklaşımların aksine kısa vadeli veriler kullanılarak ülkeler için yüksek başarımlı öngörülebilirlik tahmini yapılabildiği gösterilmiştir. Önerilen modeller KDK'ların derecelendirme sonuçlarına göre daha objektif ve sadece kısa vadeli veri gerektirmesinden dolayı daha dinamik sonuçlar sunmaktadır. Böylece yatırımcılar, KDK'ların sonuç güncellemediği aralıklarda önerilen modelleri kullanarak güncel değerler ile kendi tahminlerini üretebileceklerdir.

Sovereign credit rating is determined using quantitative and qualitative features developed by the CRAs, taking into account the economic and political risks of the countries. With the complexity of financial markets and the increase in asymmetric information, the importance and need of credit ratings have increased over time. The impact of internationalization and the widespread free market economy have removed obstacles for investors. With the increase in international financial and economic integration, country credit ratings have become one of the most important elements that direct the global capital flow. The impact of CRAs on global and national economies has increased. Over time, increased financial market complexity and borrowing diversity have gained confidence investors and regulators in the views of CRAs. Credit ratings affect investor decisions, thus influencing fund inflow and borrowing costs. Due to these factors, credit ratings play a major role in the economic development and progress of countries.In this study, the international credit rating agencies of Standard and Poor's, Moody's and Fitch, which make up the majority of the market share, are examined. Standard and Poor's to determine sovereign credit ratings assesses five key rating factors: institutional, economic, external, financial and monetary assessments. Fitch gives the sovereign credit ratings by considering four key rating factors: structural characteristics, external finance, public finance, macroeconomic performance, policies, and expectations. Moody's determines sovereign credit ratings while considering four key factors: economic and institutional power, financial strength, and sensitivity to crises. In this study, sovereign credit rating methodologies of CRAs and studies in relevant literature are examined in detail, and two dynamic methods are proposed. Credit ratings that are transparent, impartial and reliable as well as being up to date, quickly and easily calculated will provide convenience to investors and countries. The aim of this study is to develop two methods that classify countries as an alternative to long and complex sovereign credit rating methodologies modeled by Credit Rating Agencies (CRAs). These models classify countries as investable or speculative in the short term. In the first model, we used only stock market closing data from 40 composite indexes with the Artificial Neural Networks (ANNs). Ratings are determined according to short-term foreign currency. The results that we acquired from these two models are fully compliant with Standard and Poor's. However, when compared to the ratings of Fitch, the results differed in the case of Russia. In the second model, we used stock market values and macroeconomic variables with the Normalized Least Mean Square (NLMS) algorithm. Ratings for 15 countries are determined according to the short-term domestic currency. The results that we obtained from this model are fully consistent with those of Fitch. When we compared the results with Standard and Poor's, we obtained different results for Turkey and Portugal. In this first study, multilayer feed-forward networks consisting of one hidden layer were used. In the hidden layer, the hyperbolic tangent function was used as a transfer function. A linear function was used in the output layer. The hidden layer contains three cells, and the output layer contains one cell. In the network training process, Mean Square Error was used as a cost function. Gradient descent with momentum and adaptive learning rate were used to update the weight values. Changes in macroeconomic variables also affect and reflect the stock exchange market. Therefore, in this study, we only used stock exchange variables. A feed-forward network consisting of a single layer of three hidden neurons was used. For this model, 239 trading days' daily closing data until 29.08.2017 from 40 stock exchange indices of 38 countries were used. Countries from various regions of the world were included. The dataset is divided into 70% training data and 30% test data. Test data consist of the stock composite index of 12 countries in the America-Dow Jones Index, Argentina, Indonesia, Ghana, India, Hong Kong, Sweden, Malaysia, Mexico, Russia, Turkey, and Greece. In the model, credit ratings are set as '1' for investable grade and '0' for speculative grade. The stock price of the countries included in the training set have been trained according to the determined credit ratings (1 or 0). After the training, the rating classes of the countries in the test set were estimated. In the training and test results, '1' is assigned to the values that are greater than 0.6 for investable class. Values smaller than 0.4 are assigned to the speculative class by assigning a value of '0'. For the remaining values, it is planned to assign a value of '0.5'. However, when we look at training and test results as shown in Table 3.2, there is no class of '0.5' because none of the countries are between 0.4 and 0.6. If we had value in this range, we would use fuzzy logic methods to decide which one is investable or not. The model aims to classify the countries as investable or speculative with the Artificial Neural Networks method based on the sovereign credit ratings of the CRAs in terms of short-term foreign currency. The results were compared with the sovereign credit ratings of the CRAs in terms of short-term foreign currency. Table 3.2 indicates ANN results and class of the countries in the train and test set. The results that we acquired from these two models are fully compliant with Standard and Poor's. However, when compared to the ratings of Fitch, the results differed in the case of Russia. Thus, this method can be used for determining the changes in the credit ratings of countries and predicting the credit ratings of countries. The application was made with the Matlab 2016 (b) program. The Matlab code is included in Appendix C. The dataset used in the study can found in Appendix E.In this second study, we propose alternative models to sovereign credit rating methods that classify countries as investable or speculative. The model assesses countries using macroeconomic variables and stock market index monthly values with the NLMS algorithm. In this study stock composite indices, GDP per capita, U.S. foreign exchange rate, foreign trade, overnight interest rate and consumer price index values are used. The dataset includes 15 countries' monthly values between January, 2007-August, 2017. The countries in the dataset are Germany, USA, Brazil, France, Netherlands, UK, Spain, Italy, Canada, Mexico, Norway, Portugal, Russia, Turkey, and Greece. The value of the stock composite index for the end of the next month is estimated with this dataset. Therefore, the results are compared with stock composite index values between 02.2007-09.2017. Initially, the data are adjusted with the normalization equation (3.1). In normalization techniques, Min-Max Normalization is used for normalization of the data. Thus, the data are limited between 0.1 and 0.9. All data normalize in itself for it can be comparable. The proposed model consists of two phases. In the first step, stock market value is estimated by macroeconomic variables and stock market historical values using the NLMS algorithm. Macroeconomic variables were determined by considering the parameters used by the CRAs and the studies about CRAs. In the second stage, classification of countries is carried out according to the estimation results. In the NLMS model, the μ value is used as 0,12. Although convergence is faster when the selected μ value is big, convergence is slower but better if the μ value is smaller. For this reason, the μ value is determined by trial. The initial weights were determined as 0.1, which is the end result of the experiments. After the forecasting, Mean Absolute Percentage Error (MAPE) values for the last 12 months are calculated for all countries. MAPE values of countries are sorted from small to large. As seen in Fig. 3.2, there is a first biggest break in values after Spain's MAPE value. On the other hand, when we look at the derivation of MAPE function, in the first increase of the series, we see that there is a deviation after the eleventh point. Because the average of differentiation of MAPE function (T) is 1.18 to the formula (3.4), the threshold value is set at 11.8%. Therefore, if the MAPE value of a country is less than 11.8%, it can be an investable country. If the MAPE value is greater than 11.8%, the country is classified as speculative. Table 3.3 lists the stock code, MAPE values and investable and speculative classes of the countries as of October 2017. The classification for the NLMS method completely corresponds to the classification of Fitch. Additionally, compared with the Standard and Poor's classification, it was identical except for Turkey and Portugal. The application was made with the Matlab 2016 (b) program. The Matlab code for the NLMS method is included in Appendix D. Macroeconomic data in the dataset was obtained from the OECD database, stock index values and sovereign credit ratings from the Thomson Reuter Eikon program in Istanbul Technical University Finance Laboratory.In this letter, it has been shown that contrary to standard approaches, high predictability can be made for countries using short-term data. The suggested model is more objective than CRA ratings and offers more dynamic results because only short-term and public-data are required. Therefore, investors will be able to produce their own estimates with current values using the models recommended at intervals when the CRAs do not update.