Abstract
City indexes are not only a significant indicator of regional development but also a very useful guide for decision makers interested in investment to a specific region. Different city indexes have been calculated by İstanbul Stock Exchange (ISE) in order to reflect the financial performances of cities. The main purpose of this study is to predict the future behaviors of İstanbul city index which has the highest share of stocks being traded on ISE. To achieve this purpose, an important pattern recognition technique that produces reliable estimates, Hidden Markov model (HMM), is suggested. The model is constructed with four exogenous factors such as exchange rate, interest rate, money supply and consumer price index and the validity of model is shown by one-, two – and three-months ahead successful prediction results.