原文传递
Best Predictive Glmm Using Lasso with Application on High-Speed Network.
| 题名： | Best Predictive Glmm Using Lasso with Application on High-Speed Network. |
| 作者： | Hu, K.; Choi, J.; Jiang, J.; Sim, A. |
| 关键词： | Mathematics, Generalized linear mixed models, Network traffic performance, Mean squared prediction error (MSPE), Model misspecification, LASSO regularization, Computing, Computation |
| 摘要： | Efficient data access is essential for sharing massive amounts of data with many geographically distributed collaborators. Better data routing and transfers are possible for large data transfers by accurate estimations of the network traffic performance with a probabilistic tolerance. Such estimations become non-trivial when amounts of network measurement data grow in unprecedented speed and volumes, and mis-speicifed models are given. We present a statistical prediction method for network traffic performance by analyzing network traffic patterns and variation with the network conditions via the Generalized Linear Mixed Models (GLMMs), which relax the distributional assumption to that only involving the mean and variance of the errors. The method allows 鈥渂orrowing strength鈥?in the data by adopting fixed effects for shared relationship, random effects for subject variation, and errors for additional unexplained variation. The main contributions of the proposed method include: (1) best prediction accuracy even under a model misspecification; and (2) least computation time among all existing predicative algorithms under the GLMM setting. |
| 报告类型： | 科技报告 |
相关文献
- STRATEGIC ENVIRONMENTAL ASSESSMENT OF EUROPEAN HIGH-SPEED TRAIN NETWORK.
- Incorporating High Speed Passenger Rail into a Multimodal Network Model for Improved Regional Transportation Planning. NEXTRANS Project No. 055PY03.
- Texas Should Require Homeland Security Standards for High-Speed Rail.
- TRAFFIC MANAGEMENT WITH A REAL-TIME GEOGRAPHIC INFORMATION SYSTEM AND HIGH-SPEED COMMUNICATIONS NETWORK.
- High-Speed Rail and Local Land Development: Case Studies in London and Las Vegas.
- Best Practices in Shared-Use High-Speed Rail Systems.
检索历史
应用推荐
