Coffee is an extensively consumed crop grown everywhere in the world. It has become a vital agricultural crop in more than 80 tropical international locations with an anticipated 125 million human beings depending on coffee production. The purpose of this study was modeling clustered longitudinal factors on Arabica coffee bean yield at Oromia region using a linear mixed model. Data were obtained from Central Statistical Agency from January 2005 to January 2021. The study was conducted in 50 districts which were categorized into 10 zones. The minimum and maximum coffee yields were 6.77 and 7.16 Quintal/Hectare respectively. The AR(1) covariance was -4275.4 indicates a smaller value than other covariance types and the model was fitted the data well. The variables such as precipitation, soil witness, phosphorus within the soil, humidity, soil moisture, temperature and rainfall had a significant effect on coffee production via a model. The outcome of the model was accounted 84.14% variability in a district's coffee yield. Approximately 11.96% of the total variance within the district's coffee yield was due to the zone’s organizational consequences. We recommended maximizing the capability of rain-fed agriculture in the face of the threats posed by climate trade.

Keywords: Coffee Bean, Clustered Longitudinal Factors, Linear Mixed Model, Covariance Structure, Hierarchical Model