|
Title: Examining Common Distributional Assumptions of Travel Characteristics for Data Simulation
Accession Number: 01091046
Record Type: Component
Record URL: Availability: Transportation Research Board Business Office 500 Fifth Street, NW Find a library where document is available Abstract: Issues about travel data limitation have been a growing concern for many urban areas. To address this problem, new research areas regarding the transferability of household travel survey data are emerging. It has been shown that local updating can significantly improve the quality of the transferred data. As part of a comprehensive research effort on the transferability of National Household Travel Survey (NHTS) data, this study attempts to fit 13 continuous distributions and 6 discrete distributions to various travel characteristics derived from the NHTS 2001 data sets. The best-fitted distributions are selected according to multiple criteria, including Anderson–Darling, chi-square, and Kolmogorov–Smirnov tests. The results of the analysis suggest that the assumption of normality does not hold for any of the travel characteristics in the sample. Instead, distributions such as gamma, Weibull, and exponential prove to dominate the best-fitted distributions for their respective applications. Despite the complexity of the best-fitted distributions, certain stable distributional patterns are shown to exist. This paper also summarizes the best distributional assumptions for various travel characteristics, followed by a brief introduction to their applications.
Monograph Title: Monograph Accession #: 01147424
Report/Paper Numbers: 09-3443
Language: English
Authors: Zhang, YongpingMohammadian, AbolfazlPagination: pp 63-73
Publication Date: 2009
ISBN: 9780309126380
Media Type: Print
Features: Figures
(5)
; References
(19)
; Tables
(2)
TRT Terms: Identifier Terms: Uncontrolled Terms: Subject Areas: Data and Information Technology; Highways; Planning and Forecasting; I72: Traffic and Transport Planning
Files: TRIS, TRB, ATRI
Created Date: Jan 29 2008 3:47PM
More Articles from this Serial Issue:
|