The invention provides a traffic flow missing data estimation method based on the MCMC algorithm. The estimation method comprises the steps that S1) traffic flow data of N continuous days is received, a vector set of the traffic flow data is obtained according to the traffic flow data of the N continuous days, the vector set of the traffic flow data comprises observation data and missing data, and N represents a positive integer; S2) a Gaussian model is set according to the vector of the traffic flow data in the ith day; S3) the occurrence probability of the missing data is calculated according to an estimation value of a parameter space of the Gaussian model, the occurrence probability of the parameter space is calculated according to the present observation data and the newest missing data, and the estimation value of the parameter space of the Gaussian model is updated according to the occurrence probability of the parameter space; and S4) the step S3) is implemented repeatedly till Markov chain convergence is obtained, and the missing data of the traffic flow is estimated and thus, obtained. The method of the invention can greatly improve the estimation precision and speed of the missing data of the traffic flow.
Traffic flow missing data estimation method based on Markov chain Monte Carlo (MCMC) algorithm
2015-12-30
Patent
Elektronische Ressource
Englisch
Markov Chain Monte Carlo Modular Ensemble Tracking
British Library Online Contents | 2013
|Sensitivity Analysis of Markov Chain Monte Carlo
British Library Conference Proceedings | 2010
|Image Segmentation by Data Driven Markov Chain Monte Carlo
British Library Conference Proceedings | 2001
|