A Predictive Approach to Managing University Faculty Budgets in Response to Applicant Trends using Multiple Linear Regression

This study proposes a predictive modeling approach using Multiple Linear Regression (MLR) to support faculty budget planning at KH. A. Wahab Hasbullah University by forecasting student enrollment based on historical applicant trends. The urgency of this research lies in the growing need for data-driven financial planning in higher education institutions facing dynamic enrollment patterns. The dataset comprises admission records from 2013 to 2024, including applicants from Senior High Schools (SMA), Vocational High Schools (SMK), and Madrasah Aliyah (MA), for the Informatics Engineering (TI) and Information Systems (SI) programs. Six independent variables—covering applicant origin and initial university interest—were used for each program. MLR was selected due to its suitability for multivariate prediction problems and implemented using Python with the scikit-learn library. Model performance was evaluated using 10-Fold Cross-Validation, yielding an average Mean Squared Error (MSE) of 0.1951 and Mean Absolute Percentage Error (MAPE) of 0.15% for TI, and MSE of 0.3336 with MAPE of 0.47% for SI. These results demonstrate the model's strong predictive capability, particularly in the TI program. The output enables better alignment of budget allocations with expected enrollment, minimizing financial misestimations and optimizing resource distribution. This study contributes a practical and scalable framework for enrollment-based budgeting and highlights the potential of statistical modeling in institutional financial planning. The approach is adaptable for other educational institutions aiming to enhance operational efficiency through predictive analytics.