Scholarlink

<p>Business ecosystem resilience has become an important concern for policymakers and firms operating in increasingly uncertain global environments. Despite the widespread use of national resilience indices, limited research has examined how multiple macro-level institutional, social, and economic indicators jointly relate to resilience outcomes within an integrated analytical framework. This study addresses this gap by conducting a probabilistic network-based analysis of national business resilience using the 2024 FM Global Resilience Index (GRI) dataset, which covers 130 economies and 12 macro-level indicators. Rather than analyzing resilience drivers independently, the study explores how these indicators are conditionally related within a broader resilience system. A Bayesian Belief Network (BBN) framework is used to examine conditional dependencies among variables, including corruption control, education, logistics performance, political risk, and infrastructure-related indicators. The network structure is learned from observational data, and arc directions are interpreted as statistical dependencies that improve model fit rather than as causal relationships. The results indicate that education exhibits the highest mutual information with the GRI score, suggesting that variation in education levels is strongly associated with differences in resilience profiles across countries. Governance-related indicators, particularly corruption control, also display notable conditional associations with logistics performance and broader resilience states within the network. Comparisons across resilience categories further reveal systematic differences between high- and low-performing countries in indicators related to governance capacity, infrastructure quality, and social investment. By modeling resilience indicators as an interconnected probabilistic system, the study contributes to the resilience literature by providing a systems-oriented empirical assessment of macro-level determinants of resilience and demonstrating the value of BBN models for analyzing complex interdependencies among national resilience indicators. The findings offer a diagnostic analytical framework for identifying structural patterns associated with resilience while avoiding causal interpretation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Diagnostic analysis of business ecosystem resilience using Bayesian belief networks: a global competitiveness perspective

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月4日

查看全文
  • Fazeelat Aziz,
  • Abroon Qazi

摘要

Business ecosystem resilience has become an important concern for policymakers and firms operating in increasingly uncertain global environments. Despite the widespread use of national resilience indices, limited research has examined how multiple macro-level institutional, social, and economic indicators jointly relate to resilience outcomes within an integrated analytical framework. This study addresses this gap by conducting a probabilistic network-based analysis of national business resilience using the 2024 FM Global Resilience Index (GRI) dataset, which covers 130 economies and 12 macro-level indicators. Rather than analyzing resilience drivers independently, the study explores how these indicators are conditionally related within a broader resilience system. A Bayesian Belief Network (BBN) framework is used to examine conditional dependencies among variables, including corruption control, education, logistics performance, political risk, and infrastructure-related indicators. The network structure is learned from observational data, and arc directions are interpreted as statistical dependencies that improve model fit rather than as causal relationships. The results indicate that education exhibits the highest mutual information with the GRI score, suggesting that variation in education levels is strongly associated with differences in resilience profiles across countries. Governance-related indicators, particularly corruption control, also display notable conditional associations with logistics performance and broader resilience states within the network. Comparisons across resilience categories further reveal systematic differences between high- and low-performing countries in indicators related to governance capacity, infrastructure quality, and social investment. By modeling resilience indicators as an interconnected probabilistic system, the study contributes to the resilience literature by providing a systems-oriented empirical assessment of macro-level determinants of resilience and demonstrating the value of BBN models for analyzing complex interdependencies among national resilience indicators. The findings offer a diagnostic analytical framework for identifying structural patterns associated with resilience while avoiding causal interpretation.