Answers are grounded in this paper and cite their sources.
The study aims to analyze the effects of macroeconomic indicators on the Istanbul Stock Exchange (ISE 100) index using deep learning techniques based on artificial neural networks, a technology within the realm of artificial intelligence. To forecast ISE 100 index the monthly closing prices for seven macro-economic indicators (Money Supply (M2), Producer Price Index (PPI), Industrial Production Index (IPI), Exchange Rate (USD/TRY), Bond Interest Rates, Brent Oil Prices, and Gold Prices) are used in January 2001 to January 2022. The ISE 100 index has been forecasted within the framework of both deep learning models, including Long Short-Term Memory (LSTM), and Logit and Probit models, and the forecast performances of these methods have been compared. Analysis results reveal that the effects of selected macroeconomic variables on the ISE 100 index intensify during crisis periods, including the COVID-19 pandemic and the global financial crisis of 2008. In addition, the changes in the real sector have a significant impact on the capital sector, and the deep learning model is more successful in predicting financial crises. The results of the analysis show that the LSTM deep learning architectures, developed to analyze the effects of macroeconomic indicators on the ISE 100 index, have a very low error level and give more effective and satisfactory results than Logit-Probit regression models. These results offer guidance information to emerging markets and policymakers.
No takes yet. Share an insight, caveat, or question.
ÜSTÜNER et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: