Teknik Göstergeleri Makine Öğrenimi Modellerine Entegre Ederek Şehir Endeksi Hareketinin Yönünü Tahmin Etmeye Yönelik Bir Yaklaşım
An Approach to Predicting Direction of City Index Movement by Integrating Technical Indicators into Machine Learning Models

Summary/Abstract: City indexes that have been calculated by Borsa Istanbul provides a deeper understanding to investors who are interested on making investments to a specific region. Financial performances of 13 cities in Turkey are reflected by these indexes from the beginning of the year 2009. This study differs from the earlier ones mostly focused on the volatility by integrating the technical indicators into different machine learning models for the purpose of forecasting the direction of movement in the daily city indexes. The proposed procedure was applied to the Istanbul city index (XSIST) series, which has the highest number of stocks traded in BIST. 38 different technical indicators based on volume, volatility, trend and momentum were calculated and the most effective indicators in the daily change of XSIST series were selected as inputs to 6 different machine learning models. The performance of the learning models was compared with the help of metrics based on confusion matrices.

• Details
• Contents