http://ir.bdu.edu.et/handle/123456789/1858 2026-10-13T01:50:44Z 2026-10-13T01:50:44Z Anteneh, Mulugeta Eyasu http://ir.bdu.edu.et/handle/123456789/17410 2026-10-05T13:34:32Z 2026-06-01T00:00:00Z

The Impact of Crop Commercialization on Multidimensional Poverty In Rural Ethiopia: Evidence from Advanced Mixed and Causal Models Anteneh, Mulugeta Eyasu Poverty remains one of the most critical development challenges globally, particularly in low income agrarian countries such as Ethiopia, where rural livelihoods depend heavily on agriculture and multidimensional deprivation remains widespread. Although previous studies have examined the welfare effects of agricultural commercialization, most have relied on income-based poverty measures, conventional regression approaches, or geographically limited data, with limited attention to multidimensional poverty, spatial heterogeneity, and causal inference. This dissertation addresses these gaps by examining the causal impact of crop commercialization on multidimensional poverty among rural households in Ethiopia using nationally representative data from the fourth round of the Ethiopian Socioeconomic Survey (ESS 2018/19). The study analyzed 2,714 rural households nested within 59 administrative zones, integrating household, cluster, and spatial information. Multidimensional poverty was measured using the Alkire–Foster framework based on five dimensions and sixteen indicators covering nutrition & health, education, living standards, rural livelihoods & resources, and risk. A range of advanced statistical methods was employed, including generalized linear mixed models, multilevel spatial models, Targeted Maximum Likelihood Estimation (TMLE), multilevel causal mediation analysis, generalized propensity score methods for multiple treatments, and causal forest machine learning algorithms. The findings show that 47.8% of rural households were multidimensionally poor, with living standards contributing the largest share of deprivation. Poverty was disproportionately concentrated among subsistence-oriented and female-headed households, with significant spatial clustering across administrative zones. Across multiple modelling approaches, crop commercialization was consistently associated with a statistically significant reduction in multidimensional poverty. TMLE combined with machine learning and ensemble methods produced the most efficient and least biased estimates. Commercial-oriented households had a 3.56 percentage point lower probability of multidimensional poverty than subsistence households, while semi commercial households experienced smaller gains. Mediation analysis indicated that household savings partially explained the poverty-reducing effect of commercialization. Heterogeneity analysis further revealed stronger benefits among households located closer to markets and roads, with larger landholdings and younger household heads. The results suggest that crop commercialization can serve as an effective pathway for reducing rural xi multidimensional poverty, but its benefits depend on complementary investments in rural infrastructure, financial inclusion, education, and market access. By integrating causal inference, spatial statistics, and machine learning, this dissertation provides robust and policy relevant evidence to support targeted and place-based rural development strategies in Ethiopia.

2026-06-01T00:00:00Z Mesfin, Mulu http://ir.bdu.edu.et/handle/123456789/17397 2026-09-25T12:42:00Z 2026-07-01T00:00:00Z

Statio-Temporal Modeling and Deep Learning–Based Forecasting Of Severe Food Insecurity In Africa Mesfin, Mulu Food insecurity remains a major humanitarian and development challenge in African countries. Understanding its distribution, underlying determinants, and future trajectories is essential for strengthening early-warning systems and targeted interven tions. Therefore, this study examined the spatio-temporal dynamics and determinants of severe food insecurity across 52 African countries during 2015–2021 and projected its future trajectories. The dataset was sourced and compiled from food and agricul tural organization, world bank, malaria atlas project, and Ethiopian statistical service. Inference was carried out using integrated nested Laplace approximation algorithm with Gaussian Markov random fields. The analyses employed different methodolog ical approaches including spatial and spatio-temporal clustering analytical tools, ma chine learning models, spatial models, spatio-temporal models, spatio-temporal vary ing coefficient model and deep learning architectures. Among the models considered in each methodological class, extreme gradient boost, geoadditive mixed model, spatio temporal type I interaction model, and spatio-temporal graph convolutional network outperformed the other models within their respective classes. The results showed that malaria incidence (β = 0.01, SD = 0.003), climate change (β = 0.65, SD = 0.31), livestock production ( β = −0.05 , SD = 0.02), and investment inflow (β = −0.03, SD = 0.01) exhibited relatively consistent and statistically significant linear fixed ef fects across countries and over time at 95% credible intervals, whereas unemployment, greenhouse gas emissions, gross domestic product per capita, dietary energy supply, and dietary protein supply demonstrated statistically significant spatial and temporal hetero geneity in both the magnitude and, in some instances, the direction of their effects. The findings also showed that severe food insecurity was significantly spatially dependent (τ 2 θ = 2705.77) and temporally correlated (τ 2 α = 10.75) across the continent. In addition, the space-time interactions (τ 2 δ = 29,438.77) contributed additional variability. The find ings also revealed substantial and persistent disparities across African countries over time. Throughout the study period, South Sudan, Central Africa Republic, Republic of the Congo, Malawi, Guinea, Cameroon, and Somalia experienced higher severe food insecurity risks compared to other countries. The southern, eastern and central subre gions exhibited strong spatial dependence in severe food insecurity. In addition, across much of the study period, Kenya, Ethiopia, Niger, and Burkina Faso frequently expe rienced unusual country-year-specific shocks in severe food insecurity crises compared to other countries. Data-driven forecasts further suggested that severe food insecurity is expected to intensify in Africa. In particular, severe deterioration was projected in South Sudan, the Central African Republic, the Republic of Congo, Malawi, Guinea, Somalia, viii Ethiopia, Kenya, Democratic Republic of Congo, Angola, Namibia, Niger and Nigeria. These findings indicated the coexistence of cross-country shared structural influences and country-specific factors in shaping severe food insecurity across Africa. Policy makers should prioritize targeted and context-specific interventions in high-risk areas, strengthen national multi-sectoral and regional coordination, and emphasize proactive and anticipatory strategies, including early warning systems, to support progress toward achieving sustainable development goal 2 (Zero Hunger) across Africa.

2026-07-01T00:00:00Z Mulugeta, Tesfa http://ir.bdu.edu.et/handle/123456789/17064 2026-08-21T07:02:31Z 2026-06-01T00:00:00Z

Comprehensive Statistical Analytics for Sorghum Genotype Selection Under Irrigated and Non-Irrigated Conditions Mulugeta, Tesfa Sorghum is a widely cultivated crop in arid and semi-arid regions, particularly in sub-Saharan African countries. Several factors have been influencing sorghum production, with genotype and genotype-by-environment interaction being the primary sources of variability in yield and yield-related traits. This dissertation aimed to identify the best statistical approach for sorghum genotype selection with high performance in grain yield and yield-related traits under non-irrigated and irrigated conditions. Specifically, the sorghum genotype selection with superior performance in grain yield under both conditions is done by treating the genotypes as fixed or random effects using parametric, nonparametric statistical models and unsupervised machine learning to perform the genotype selection with superior performance in major traits. The experiment used water availability as a treatment, and each replication within the treatment levels used a lattice square design for data collection. A design consisted of 14×14 square experimental units containing 196 genotypes, where each row of the square represented a block receiving 14 genotypes, from which phenotypic traits were measured for the analysis. Findings revealed that the sorghum genotypes with superior performance in grain yield employed different parametric models. A result of genotype selection using mean performance in analysis of variance, with an assumption of genotypes treated as fixed effects, which is less important compared to other parametric models considering the genotypes as random effects. The genotype selection with superior performance in grain yield using mixed-effects models was a more reliable selection compared with the classical model. Findings of genotype selection with superior performance in major traits using a distributional assumption-free statistical model were advisable to scan the genotypes performing best in various phenotypic traits. Classification of phenotypic traits and sorghum genotypes was implemented to investigate high-performing genotypes in major phenotypic traits. The outcomes of this study can help academics and researchers in plant breeding for sorghum genotype selection in phenotype traits using advanced models. This can serve as an alternative statistical analysis approach for plant breeders.

2026-06-01T00:00:00Z Getnet, Bogale http://ir.bdu.edu.et/handle/123456789/16991 2026-08-18T06:08:52Z 2026-06-01T00:00:00Z

A Probabilistic and Data-Driven Approach to Childhood Under nutrition Transitions Getnet, Bogale Child undernutrition remains a persistent public health challenge in Ethiopia, contributing to significant developmental, health, and economic burdens. Despite national commitments to reduce its prevalence, rates of stunting, underweight, and wasting remain high, particularly among children in rural and low-income households. Previous studies have often examined undernutrition in isolation, without accounting for its concurrent forms or the complex pathways through which it evolves over time. This study aimed to analyze the dynamic nature of childhood undernutrition in Ethiopia, focusing on the transitions between nutritional states and the combined effects of social, economic, environmental, and maternal factors. The research also investigates which factors predict long-term nutritional outcomes and how interventions can be better targeted to high-risk populations. Methodologically, the research applied advanced statistical and computational tools to longitudinal panel data, including multistate Markov chain models for state transitions, dynamic Bayesian networks for causal exploration, machine learning, and deep learning techniques for long-term trend prediction. A sample of over 1,999 children, aged 1 to 15, was followed across five survey rounds (2002–2016), with data collected from five diverse regions in Ethiopia-capturing a broad representation of geographic and socioeconomic contexts. The findings revealed that undernutrition is frequently experienced in multiple forms simultaneously and is influenced by a constellation of interrelated factors-particularly maternal education, household wealth, and access to health services and sanitation. Results also showed that many children transitioned between undernutrition states over time, with some remaining chronically affected. Predictive models accurately identified high-risk groups and highlighted early adolescence as a critical window for intervention. These results offer valuable insights for Ethiopian health planners and policymakers, reinforcing the need for integrated, long-term, and equity-focused strategies. By uncovering dynamic patterns and causal pathways, this study contributes to both academic understanding and practical decision-making for child nutrition programs in Ethiopia and similar contexts

2026-06-01T00:00:00Z