General Regression Neural Networks in Forecasting the Scales of Higher Education

The historical scales of higher education of a given area can be viewed as a time series which is charactered by uncertainty, nonlinearity and time-varying behavior. Predictions for the number of enrolled students in colleges of Shandong province of China and its modified data were carried out respectively by means of General Regression Neural Network (GRNN) forecasters. The detailed designs for architectures of GRNN models, transfer functions of the hidden layer nodes, input vectors and output vectors, were made with many tests. Experimental results show that the performance of GRNN for forecasting the scales of the near future scales of higher education is acceptable and effective.