O.R. Applications A Bayesian causal map for inflation analysis: The case of Turkey S ß ule O ¨ nsel S ß ahin a , Fu ¨sun U ¨ lengin a, * , Burc ¸U ¨ lengin b a Dogus University, Acibadem, 34722 Kadikoy, Istanbul, Turkey b Istanbul Technical University, Management Faculty, Industrial Engineering Department, 80680 Macka, Istanbul, Turkey Received 25 August 2003; accepted 27 June 2005 Available online 24 August 2005 Abstract This paper proposes the use of Bayesian Causal Maps (BCMÕs) to analyze the complex structure of inflation in Tur- key. In this study, a model of inflation is initially structured using a cognitive mapping technique; the dependent prob- abilities of the concepts are then calculated based on the detailed analysis of past data. Finally, a BCM is used to analyze the complex structure of inflation in Turkey. As a result, it will be possible to see the structure of the inflation model and to understand the basic consequences of any strategic change that may occur in the system. Ó 2005 Elsevier B.V. All rights reserved. Keywords: Economics; Forecasting; Cognitive mapping; Bayesian causal map; Turkish Inflation 1. Introduction All traditional forecasting methods are based on the assumption that tomorrowÕs world will be much like todayÕs(Le Bihan and Sedillot, 2000; O ¨ zatay, 2001; Stock and Watson, 1999; Ulengin et al., 2000). However, due to the high level of uncertainty concerning the future, a prediction of inflation through traditional forecasting is insufficient in understanding the real dynamics of inflation or in anticipating the major shifts that may occur in an uncertain macroeconomic environment. Growth and inflation have always been among the most widely analyzed topics as well as being amongst those which have attracted the interest of macroeconomic forecast researchers (Hendry, 2001; Stock and Watson, 1999; De Brouwer and Ericson, 1998; Bonerjee and Russell, 2001). These two concepts constitute the general performance indicators of macroeconomics and, thus, macro- economic policies are configured mostly based on 0377-2217/$ - see front matter Ó 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.ejor.2005.06.036 * Corresponding author. Tel.: +90 216 32711 04/1354; fax: +90 216 327 96 31. E-mail addresses: sonsel@dogus.edu.tr (S ß.O ¨ .S ßahin), fulen- gin@dogus.edu.tr (F. U ¨ lengin), ulenginbur@itu.edu.tr (B. U ¨ lengin). European Journal of Operational Research 175 (2006) 1268–1284 www.elsevier.com/locate/ejor