Journal of System Simulation ›› 2017, Vol. 29 ›› Issue (9): 2128-2134.doi: 10.16182/j.issn1004731x.joss.201709034

Previous Articles     Next Articles

Research of State Vector in Short-Term Passengers Flow Forecasting Based on Nonparametric Regression

Guo Han, Jiao Pengpeng   

  1. School of Civil and Transportation Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, China
  • Received:2017-05-19 Published:2020-06-02

Abstract: KNNR (K Nearest Neighbor Based Nonparametric Regression) Method was used for short-term traffic forecast and the choice of state vector was studied. The result shows that taking the data of some historical periods as the state vector has a good prediction. Although the correlation of the historical passenger flow between different Rail transit sites is significant, it neglects the fact that the passengers enter each station is independent. So taking the historical passenger flow of adjacent sites as the state vector is not appropriate.

Key words: short-term passengers flow forecasting, nonparametric regression, state vector, k nearest neighbors

CLC Number: