BANKING SECTOR DOMINANCE AND CROSS-SECTORAL DEPENDENCIES IN THE IRAQI STOCK EXCHANGE: A MACHINE LEARNING AND NETWORK ANALYSIS APPROACH

Mohammed Faez Hasan (), Noor Sabah Hameed Ai-Dahaan () and Hussein Hadi Abdulameer ()
Additional contact information
Mohammed Faez Hasan: Department of Finance and Banking Sciences. University of Kerbala, Iraq
Noor Sabah Hameed Ai-Dahaan: Department of Finance and Banking Sciences. University of Kerbala, Iraq
Hussein Hadi Abdulameer: Department of Finance and Banking Sciences. University of Kerbala, Iraq

European Journal of Accounting, Finance & Business, 2025, vol. 13, issue 1, 154-164

Abstract: This study investigates the behavior of different economic sectors, the accuracy of forecasting models, and the interlocking relationships among firms in the Iraqi Stock Exchange 60-index (ISX60)-an effort to fill a major gap in scholarship on emerging markets. Using daily closing prices from 57 firms drawn from seven sectors between January 2015 and December 2024, we combine sector-contribution decomposition, machine-learning and time-series forecasting, and graph-based network analysis into a unified empirical framework. The results reveal a heavily concentrated structure, with the banking segment alone holding nearly 30 percent of total capitalization, followed by industry at 18 percent and services at 15 percent, underscoring both the prominence of financial intermediaries in the country’s growth agenda and the systemic exposures they generate. Forecasting accuracy varies sharply across sectors, and three benchmark models-ARIMA, long short-term memory (LSTM) networks, and random forests-yield distinctly different outcomes. LSTM dominates in highly turbulent groups, achieving a mean absolute error of 0.15 for banks and 0.08 for industry, and shows an ability to learn non-linear, long-horizon dependency. Random forests shine in intermediate-volatility categories, whereas ARIMA still holds ground only in calm, low-dispersion arenas. Network metrics confirm a tight coupling between banking and manufacturing (Pearson r equals 0.8), suggesting that shocks originating in financial losses could spread quickly across key productive nodes during downturns. The evidence presented here enhances the understanding of portfolio diversification, showing that agriculture and insurance can act as effective hedges because their returns move independently of larger sectors. For regulators, the results highlight the need to supervise tightly interlinked industries and, at the same time, to bolster modest-performing fields that matter for long-run stability. By focusing on Middle Eastern emerging markets, this study adds to scarce regional literature and provides actionable insights for investors, policymakers, and analysts who work in frontier-market settings.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.accounting-management.ro/getpdf.php?paperid=37_17 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:scm:ejafbu:v:13:y:2025:i:1:p:154-164

DOI: 10.4316/EJAFB.2025.13116

Access Statistics for this article

European Journal of Accounting, Finance & Business is currently edited by Stefan cel Mare University of Suceava, West University of Timisoara

More articles in European Journal of Accounting, Finance & Business from "Stefan cel Mare" University of Suceava, Romania - Faculty of Economics and Public Administration, West University of Timisoara, Romania - Faculty of Economics and Business Administration Contact information at EDIRC.
Bibliographic data for series maintained by Liviu Scutariu ( this e-mail address is bad, please contact ).