Mohammed Muntaka Abdul Rahman
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Mohammed Muntaka Abdul Rahman has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Oluwaseun Adeoye Oyebamiji
Department of Agricultural Economics, Faculty of Agriculture, Ege University, Izmir, Türkiye
2 shared publications - Asad Ul Islam Khan
Ibn Haldun University
1 shared publication - Hasan Vergil
Ibn Haldun University
1 shared publication Jones Ntiamoah
1 shared publicationKhotimatul Barki
1 shared publicationPeter Arhenful
1 shared publication
Latest Publications
- Open Access
- 14 September 2026
- Published by Elsevier in SSRN Electronic Journal
Spatial and Household Predictors of Climate Impacts: A Comparative Machine Learning Study in Indonesian Coastal Farming Communities
Climate vulnerability frameworks typically assume that household-level characteristics determine impact outcomes, requiring extensive data collection on demographics, assets, and adaptive capacity. We test this assumption using machine learning to analyze vulnerability patterns among 498 coastal rice farmers across four subdistricts in Demak, Indonesia. Applying logistic regression, Random Forest, and Leave-One-Region-Out spatial cross-validation, we discover perfect geographic determination of vulnerability: location alone achieves 100% classification accuracy (Cramér's V= 1.0, chi-square= 500, p<0.001). All farmers in inland subdistricts (Guntur, Mranggen) experienced low impacts, while all farmers in coastal subdistricts (Karangtengah, Sayung) experienced high impacts. This spatial separation is driven by unavoidable tidal flooding exposure affecting 99.6% of coastal farmers but 0% of inland farmers. Despite extensive household-level adaptation (99% embankment construction, 96% income diversification in coastal areas), impacts remain universally high, demonstrating structural constraints that overwhelm individual agency. Within high-impact regions, Random Forest regression explains 53% of continuous impact variance, with land size as the dominant predictor (importance=19.8), while low-impact regions show no predictive variation. Behavioral economics measures differ minimally between regions (largest gap 0.09 on 5-point scales) and contribute negligibly to outcomes. These findings reveal that when geographic factors create sharp exposure boundaries, the 27 household variables tested become redundant for binary classification. A simple geographic rule (if coastal then high-impact, else low-impact) eliminates the need for household surveys costing $7,500-$15,000, achieving perfect targeting with zero inclusion or exclusion errors. Results challenge individual-level vulnerability assessment paradigms and demonstrate boundary conditions where geographic targeting dominates household screening.
- 14 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessSpatial and Household Predictors of Climate Impacts: A Comparative Machine Learning Study in Indonesian Coastal Farming Communities
Climate vulnerability frameworks typically assume that household-level characteristics determine impact outcomes, requiring extensive data collection on demographics, assets, and adaptive capacity. We test this assumption using machine learning to analyze vulnerability patterns among 498 coastal rice farmers across four subdistricts in Demak, Indonesia. Applying logistic regression, Random Forest, and Leave-One-Region-Out spatial cross-validation, we discover perfect geographic determination of vulnerability: location alone achieves 100% classification accuracy (Cramér's V= 1.0, chi-square= 500, p<0.001). All farmers in inland subdistricts (Guntur, Mranggen) experienced low impacts, while all farmers in coastal subdistricts (Karangtengah, Sayung) experienced high impacts. This spatial separation is driven by unavoidable tidal flooding exposure affecting 99.6% of coastal farmers but 0% of inland farmers. Despite extensive household-level adaptation (99% embankment construction, 96% income diversification in coastal areas), impacts remain universally high, demonstrating structural constraints that overwhelm individual agency. Within high-impact regions, Random Forest regression explains 53% of continuous impact variance, with land size as the dominant predictor (importance=19.8), while low-impact regions show no predictive variation. Behavioral economics measures differ minimally between regions (largest gap 0.09 on 5-point scales) and contribute negligibly to outcomes. These findings reveal that when geographic factors create sharp exposure boundaries, the 27 household variables tested become redundant for binary classification. A simple geographic rule (if coastal then high-impact, else low-impact) eliminates the need for household surveys costing $7,500-$15,000, achieving perfect targeting with zero inclusion or exclusion errors. Results challenge individual-level vulnerability assessment paradigms and demonstrate boundary conditions where geographic targeting dominates household screening.
- Open Access
- 3 August 2026
- Published by Elsevier in SSRN Electronic Journal
Machine Learning, Return Predictability, and Implementable Alpha on the Ghana Stock Exchange
This paper examines the return predictability and portfolio performance of machine learning models relative to conventional asset pricing models on the Ghana Stock Exchange using monthly equity, macroeconomic, and commodity price data from 2007:M12-2023:M4. Results show that Elastic Net delivers the strongest out-of-sample performance, achieving an R² of 28.23%, an annualised long-short portfolio return of 88.4%, and a gross Sharpe ratio of 3.28. This substantially outperforms CAPM (-0.08% return; Sharpe-0.003) and ARIMA (14.7% return; Sharpe 0.59). Although the market consists of a concentrated eight-stock liquid universe typical of frontier markets, the performance differential remains economically and statistically significant. After accounting for GSE-specific transaction costs of 1.5%-2.5% round-trip, net Sharpe ratios remain between 1.86 and 2.96. Subsample results further indicate improving predictive performance over time, implying that regularised nonlinear machine learning models can extract persistent alpha in emerging equity markets.
- 3 August 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessMachine Learning, Return Predictability, and Implementable Alpha on the Ghana Stock Exchange
This paper examines the return predictability and portfolio performance of machine learning models relative to conventional asset pricing models on the Ghana Stock Exchange using monthly equity, macroeconomic, and commodity price data from 2007:M12-2023:M4. Results show that Elastic Net delivers the strongest out-of-sample performance, achieving an R² of 28.23%, an annualised long-short portfolio return of 88.4%, and a gross Sharpe ratio of 3.28. This substantially outperforms CAPM (-0.08% return; Sharpe-0.003) and ARIMA (14.7% return; Sharpe 0.59). Although the market consists of a concentrated eight-stock liquid universe typical of frontier markets, the performance differential remains economically and statistically significant. After accounting for GSE-specific transaction costs of 1.5%-2.5% round-trip, net Sharpe ratios remain between 1.86 and 2.96. Subsample results further indicate improving predictive performance over time, implying that regularised nonlinear machine learning models can extract persistent alpha in emerging equity markets.
- Open Access
- 8 July 2025
- Published by Firenze University Press in Journal of Agriculture and Environment for International Development (JAEID)
- Vol. 119 (1) , 413-438
Reducing Food Insecurity in Sub-Saharan Africa: The Role of Institutions and Financial Stability
Achieving food security (SGD 2) is a central theme in promoting overall well-being, economic growth, and maintaining peace and security. In the efforts to achieve sustainable development goals, Sub-Saharan African countries are still burdened with widespread food insecurity, poverty, and inequality. These bottlenecks are further worsened by institutional inefficiencies and unstable financial systems. This study investigates not only the role institutions and stable financial systems play in the reduction of food insecurity and its four components but also how poverty and well-being in SSA affect food security. The study employed data from 28 SSA countries from 2011 to 2021 which was analyzed using a two-step system GMM estimator. The results show that although financial stability is insignificant relative to food security and utilization of food, it increases the affordability and sustainability of food. Quality institution on the other hand significantly increases food security. Furthermore, well-being increases food security, affordability, utilization, and sustainability, while poverty reduces food security, accessibility, utilization, and sustainability. The study suggests that strengthening financial systems to provide financial access and security will reduce poverty and support the achievement of SDG 2. Additionally, policies to improve people’s well-being and further the existing institutions will provide a stronger foundation for SSA countries to achieve SDG 2.
- 8 July 2025
- Published by Firenze University Press in Journal of Agriculture and Environment for International Development (JAEID)
- Vol. 119 (1) , 413-438
Open AccessReducing Food Insecurity in Sub-Saharan Africa: The Role of Institutions and Financial Stability
Achieving food security (SGD 2) is a central theme in promoting overall well-being, economic growth, and maintaining peace and security. In the efforts to achieve sustainable development goals, Sub-Saharan African countries are still burdened with widespread food insecurity, poverty, and inequality. These bottlenecks are further worsened by institutional inefficiencies and unstable financial systems. This study investigates not only the role institutions and stable financial systems play in the reduction of food insecurity and its four components but also how poverty and well-being in SSA affect food security. The study employed data from 28 SSA countries from 2011 to 2021 which was analyzed using a two-step system GMM estimator. The results show that although financial stability is insignificant relative to food security and utilization of food, it increases the affordability and sustainability of food. Quality institution on the other hand significantly increases food security. Furthermore, well-being increases food security, affordability, utilization, and sustainability, while poverty reduces food security, accessibility, utilization, and sustainability. The study suggests that strengthening financial systems to provide financial access and security will reduce poverty and support the achievement of SDG 2. Additionally, policies to improve people’s well-being and further the existing institutions will provide a stronger foundation for SSA countries to achieve SDG 2.
- Open Access
- 29 February 2024
- Published by Center for Open Science in OSF Preprints
Artificial Intelligence Tools for Academic Work
The repercussions of artificial intelligence implementations for edu- cation can be observed at both an individual and collective level. At the personal level, AI can recommend suitable learning materials to students. In contrast, at the collective level, it can affect the relationships between various stakeholders, such as how educators perceive learners’ progress. Artificial Intelligence (AI) can be categorized based on its competencies, specifically Artificial Narrow Intelligence (ANI) and Artificial General In- telligence (AGI). The former, also known as weak AI, refers to the AI tools that have been developed and implemented to date. Conversely, the latter pertains to AI tools that have not yet been accomplished and, if achieved, would be comparable to human intelligence and capabilities.
- 29 February 2024
- Published by Center for Open Science in OSF Preprints
Open AccessArtificial Intelligence Tools for Academic Work
The repercussions of artificial intelligence implementations for edu- cation can be observed at both an individual and collective level. At the personal level, AI can recommend suitable learning materials to students. In contrast, at the collective level, it can affect the relationships between various stakeholders, such as how educators perceive learners’ progress. Artificial Intelligence (AI) can be categorized based on its competencies, specifically Artificial Narrow Intelligence (ANI) and Artificial General In- telligence (AGI). The former, also known as weak AI, refers to the AI tools that have been developed and implemented to date. Conversely, the latter pertains to AI tools that have not yet been accomplished and, if achieved, would be comparable to human intelligence and capabilities.