Structural signals of speculative technology cycles: a retrospective multidomain diagnostic for the MENA region
4 affiliations
- 1College of Computing and IT, University of Doha for Science and Technology, Doha, 24449, Qatar; Corresponding author.
- 2Institute for Professional and Continuing Education, Zayed University, Al Thuraya Street Academic City, Dubai, United Arab Emirates
- 3Information Systems and Technology Management Department, College of Technological Innovation, Zayed University, Dubai, 19282, United Arab Emirates
- 4School of Computer Science, University of Windsor, 401 Sunset Ave, Windsor, Ontario, N9B 3P4, Canada
Abstract
This study develops a foresight-oriented structural diagnostic for detecting speculative technology dynamics in emerging digital economies, focusing on the Middle East and North Africa (MENA) region. Moving beyond conventional macroeconomic monitoring, the framework integrates long-run structural relationships across five domains: economic fundamentals, labor markets, startup ecosystems, policy governance, and behavioral indicators. Only three met the stationarity and coverage requirements for equilibrium modelling; the exclusion rule and its sensitivity to aggregation are stated explicitly. Using 26 series from 24 macroeconomic indicators over 2000 to 2022, Johansen cointegration analysis identifies long-run equilibria within each domain. These are estimated on a regional aggregate of 23 annual observations, with one lagged difference and an unrestricted constant, under a finite-sample correction. Deviations from equilibrium, captured as error-correction terms (ECTs), signal structural stress. They are combined into a composite Speculative Technology Regime Index distinguishing build-up, overheating, and collapse phases. Logistic Regression, Random Forest, and XGBoost models then test whether these signals predict regime transitions. The index co-moves retrospectively with the dot-com episode and a mid-2010s turning point. However, it shows no predictive generalization: all three models misclassify both held-out observations. The study contributes a hybrid econometric–machine learning approach to technology foresight and socio-technical transitions research, offering policymakers a retrospective lens on structural stress. Prospective early warning remains unresolved. Achieving it would require panel-level estimation, independent regime dating, real-time signal construction, and rolling-origin validation against an explicit benchmark.
Related
- The impact of assisted reproductive technology on the neonatal prognosis of newborns < 29 weeks gestational age: a retrospective case-control study
- Integrating green building technology and renewable energy for a carbon-neutral built environment: Exploring the mechanisms of energy efficiency, AI technology, and environmental policy stringency
- CRISPR Technology Applied to the Analysis of Antibiotic Resistance Genes in the Environment.
- Technology readiness and energy transitions: Evidence from renewable energy finance and institutional quality
- How much climate-smart agriculture is enough? A dose-response analysis of technology adoption effects on farmer welfare in Ghana
BibTeX
@article{khan2026structural,
title = {{Structural signals of speculative technology cycles: a retrospective multidomain diagnostic for the MENA region}},
author = {Shafaq Khan and Munir Majdalawieh and Maher Alaraj and Mohammad Saqib Shah},
journal = {Sustainable Futures},
year = {2026},
month = oct,
doi = {10.1016/j.sftr.2026.102236},
url = {https://doi.org/10.1016/j.sftr.2026.102236},
note = {CC BY},
}
RIS
TY - JOUR
TI - Structural signals of speculative technology cycles: a retrospective multidomain diagnostic for the MENA region
AU - Shafaq Khan
AU - Munir Majdalawieh
AU - Maher Alaraj
AU - Mohammad Saqib Shah
PY - 2026
DA - 2026/10/30
JO - Sustainable Futures
DO - 10.1016/j.sftr.2026.102236
UR - https://doi.org/10.1016/j.sftr.2026.102236
AB - This study develops a foresight-oriented structural diagnostic for detecting speculative technology dynamics in emerging digital economies, focusing on the Middle East and North Africa (MENA) region. Moving beyond conventional macroeconomic monitoring, the framework integrates long-run structural relationships across five domains: economic fundamentals, labor markets, startup ecosystems, policy governance, and behavioral indicators. Only three met the stationarity and coverage requirements for equilibrium modelling; the exclusion rule and its sensitivity to aggregation are stated explicitly. Using 26 series from 24 macroeconomic indicators over 2000 to 2022, Johansen cointegration analysis identifies long-run equilibria within each domain. These are estimated on a regional aggregate of 23 annual observations, with one lagged difference and an unrestricted constant, under a finite-sample correction. Deviations from equilibrium, captured as error-correction terms (ECTs), signal structural stress. They are combined into a composite Speculative Technology Regime Index distinguishing build-up, overheating, and collapse phases. Logistic Regression, Random Forest, and XGBoost models then test whether these signals predict regime transitions. The index co-moves retrospectively with the dot-com episode and a mid-2010s turning point. However, it shows no predictive generalization: all three models misclassify both held-out observations. The study contributes a hybrid econometric–machine learning approach to technology foresight and socio-technical transitions research, offering policymakers a retrospective lens on structural stress. Prospective early warning remains unresolved. Achieving it would require panel-level estimation, independent regime dating, real-time signal construction, and rolling-origin validation against an explicit benchmark.
SN - 2666-1888
ER -