Student Enrollment and Teacher Statistics Forecasting Based on Time-Series Analysis.

Area of Science:

• Education policy
• Demographic trends
• Time-series analysis

Background:

• National competitiveness is linked to education, requiring enhanced student and teacher potential.
• Declining birthrates in developed nations impact international education competitiveness.
• Taiwan faces educational challenges due to low birthrates, decreased enrollment, and teacher surplus.

Purpose of the Study:

• To forecast student and teacher number trends in Taiwan.
• To address the impact of demographic shifts on the education sector.
• To improve the accuracy of educational time-series forecasting.

Main Methods:

• A hybrid Whale Optimization Algorithm (WOA) and Support Vector Regression (WOASVR) model was developed.
• WOA was utilized to optimize support vector kernel parameters.
• Ministry of Education data from 1991-2018 on student and teacher numbers were analyzed.

Main Results:

• Student and teacher numbers showed an annual decrease, with private primary schools being an exception.
• The WOASVR model demonstrated superior forecasting accuracy compared to other common models.
• WOASVR achieved the lowest Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE).

Conclusions:

• The WOASVR method provides accurate forecasts for student and teacher demographics.
• Accurate forecasting is essential for developing effective education policies.
• The study offers valuable insights for addressing educational challenges posed by declining birthrates.