Mud crab weight prediction using Multiple Linear Regression model

This study explores the use of multiple linear regression (MLR) to predict the weight of mud crabs (Scylla spp.) based on gender, carapace length (CL), sternum width (SW), and propodus width (PW). Data from 60 mud crabs comprising two species, Scylla paramamosain and Scylla olivacea, were analyzed. Key predictors, including CL, SW, and PW, demonstrated significant correlations with body weight, explaining 95.9% of the variance in the final regression model. Model validation showed high reliability, with adherence to linearity between the variables, normality, homoscedasticity, and error terms assumptions. The findings offer a practical tool for fisheries management, allowing efficient and non-invasive weight estimation, which aids in sustainable harvesting and population assessments. Despite the model’s strong predictive capabilities, the study acknowledges limitations, such as reliance on secondary data and geographic constraints, and suggests the inclusion of additional variables and expanded datasets to enhance generalizability. This research provides a foundation for improved aquaculture practices and the development of adaptive management strategies in dynamic environmental conditions.