Akademik Veri Yönetim Sistemi
Tezin Türü: Yüksek Lisans
Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Uygulamalı Matematik Enstitüsü, Aktüerya Bilimleri Anabilim Dalı, Türkiye
Tezin Onay Tarihi: 2015
Öğrenci: DİLEK AYDOĞAN
Danışman: SEVTAP AYŞE KESTEL
Özet:Inflation is a significant issue that is cared by every segment of society but especially by economists since it plays an important role in the economic problems of countries. In Turkey, inflation has been a substantial problem since 1970s. There is no stability at inflation rates from that time to 2003. The Central Bank of the Republic of Turkey became independent and started to apply implicit inflation target regime at 2001. As a result of these improvements inflation was taken under control at 2004. Although the rates are more stable after that time, there are some fluctuations still. Hence, it is essential to study on the inflation data of 2004-2014 year interval. In this study, a novel approach is preferred in order to analyze, model and predict the inflation data of Turkey for the mentioned period. Hidden Markov Models are applied in several fields but there are few applications of it to analyze inflation. Now, theoretical background of HMMs will be provided. Then, the advantage of performing well on autocorrelated data just like ours of the model will be taken. The monthly inflation rates that represent the price level changes at current month with respect to the same month of previous year will be accepted as observed part of the model and the hidden part is estimated by EM-algorithm approach adapted to normal-HMM. The first stage model will be applied to the data for the years 2004-2012 and the test period of 2013-2014.