Forecasting an economic growth model using artificial neural networks and application of Turkey

MUSTAFA İSA DOĞAN

Forecasting an economic growth model using artificial neural networks and application of Turkey

Thesis Type: Postgraduate

Institution Of The Thesis: Gazi Üniversitesi, Fen Bilimleri Enstitüsü, Turkey

Approval Date: 2012

Student: MUSTAFA İSA DOĞAN

Supervisor: HACI HASAN ÖRKCÜ

Open Archive Collection: AVESIS Open Access Collection

Abstract:

Forecasting the economic growth is very difficult because of the crisis fluctuations in economy. Especially the classical linear regression models and time series models can not turn out satisfactory in these kind economic regression indicators. There can be many reasons of that; however the most over-riding reason is the data structures are non-linear. Because of that, in that kind of cases, we should apply the Artificial Neural Networks which is a non-linear model, for receiving the most accurate results. Economic growth is defined as the increase of the average per capita income. Our economic life standards are depending to the growth in gross domestic product (GDP). The purpose of this study is to obtain information about the general situation of the economic growth in Turkey according to quarterly data between the years 1999 to 2011 and using Artificial Neural Networks (ANN) model in forecasting the economic growth. For that reason, the comparison between ANN model and Linear Regression Analysis had been made in the context of Root Mean Square Error. The results proved that ANN is an effective tool to forecast the economic growth.