Research Article

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

Table 4

Performance comparison of different forecasting models for teacher dataset.

Number of teachersCriteriaARIMAETSTBATSGRIDSVRPSOSVRWOASVR

Primary schoolPublicMAPE (%)1.651.672.752.121.461.39
RMSE1835.081915.493259.282522.981769.121702.23
PrivateMAPE (%)3.838.1510.886.923.442.86
RMSE84.62159.46236.42158.9277.2161.8

Secondary schoolPublicMAPE (%)2.504.276.544.602.252.04
RMSE1427.812348.593901.302665.911340.901306.95
PrivateMAPE (%)5.576.2613.0312.424.613.93
RMSE39.0641.0668.3087.8838.5932.67

High schoolPublicMAPE (%)1.132.481.402.550.70.69
RMSE474.531058.47675.91953.11292.2288.08
PrivateMAPE (%)3.274.104.383.913.093.07
RMSE757.49906.791094.97815.19712.99702.78

UniversityPublicMAPE (%)2.451.542.651.591.311.21
RMSE931.06600.89999.42715.22527.44471.26
PrivateMAPE (%)2.423.2218.084.762.361.58
RMSE1614.252379.3911928.493482.681773.461376.06

AveragePublicMAPE (%)1.932.493.342.721.431.33
RMSE1167.121480.862208.981714.31982.42942.13
PrivateMAPE (%)3.775.4311.597.003.382.86
RMSE623.86871.683332.051136.17650.56543.33

ARIMA, autoregressive integrated moving average; ETS, exponential smoothing; TBATS, Trigonometric Seasonal Box–Cox Transformation with ARMA residuals Trend and Seasonal Components; GRIDSVR, grid search support vector regression; PSOSVR, particle swarm optimization support vector regression; WOASVR, whale optimization algorithm support vector regression; boldface, the best values in each row.