eng Science and Education Publishing American Journal of Applied Mathematics and Statistics 2328-7292 2025-06-26 13 2 38 46 10.12691/ajams-13-2-3 AJAMS20251323 article William Kofi Nkegbe william.nkegbe@uds.edu.gh 1 Abukari Alhassan 1 Alhassan Faisal 1 Department of Statistics, University for Development Studies, P. O. Box 1350, Tamale, Ghana Statistical modeling is fundamental in understanding and predicting the variability of climate change and its effect on cereal crop yield in Northern Ghana. Most often, there is little or no data to study these relationships within the Northern Ghana enclave. The Dynamic Bayesian Network Modeling was employed to model the relationship using data on some selected cereal crops (Maize, Rice, and Sorghum) yield from Ghana¡¯s Ministry of Food and Agriculture, and some climate variables (rainfall, sunshine, temperature, humidity, and windspeed) from Ghana Meteorological Agency, for Northern Ghana for a thirty-one-year period. The results showed that climate variables are significantly related to yield at the intra-slide and inter-slide levels and that there exist positive and significant (link strength values > 1) relationships between cereal crop yield and climate change variables. The results imply that, with data on current climate variables under the study, the next cereal crop yield can be predicted with reduced uncertainty and better forecasting ability while holding other factors constant. On the basis of the analysis, the study affirmed the use of the Dynamic Bayesian Networks and highly recommended the extension of the study to include other variables such as vegetative cover, cultivated area, and chemical use, when data becomes available. https://pubs.sciepub.com/ajams/13/2/3/ajams-13-2-3.pdf Dynamic Bayesian Networks Climate Change Link Strength Cereal Crop Yield Northern Ghana Agriculture