论文部分内容阅读
The traditional prediction methods of the passenger traffic are generally based on the existing data to predict future passenger traffic.In this paper, we use a new way, which reflects the relationship between the passenger traffic and the factors affecting the passenger traffic, to predict the future changes of the passenger traffic according to the changes of the factors.Firstly, this paper uses the Gray Correlation Analysis method to calculate the Gray Correlation of each factor to the passenger traffic.According to the Gray Correlation Calculation, we choose 7 indexes with a high gray correlation to the passenger traffic to be the basic indicators.Then, we use the 7 indexes and the passenger traffic data to form the research data.The data from the year 1990 to 1999 is regarded to be the training data.The data from the year 2000 to 2009 is regarded as the test data.We use the train data to build the Generalized Regression Neural Network (GRNN) and forecast the passenger traffic from the year 2000 to 2009.Finally, we compare the predicted results with the actual data.The results show that the error rate belongs to [-2%, +2%].This shows that the GRNN neural network prediction algorithm has higher prediction accuracy and better prediction results.