Federal University of Technology, Minna Institutional Repository >
School of Environmental Technology >
Urban and Regional Planning >
Please use this identifier to cite or link to this item:
|Title: ||On the Effects of Outliers on Traffic Modelling Accuracy Using Multiple regression Analysis|
|Authors: ||Morenikeji, Wole|
|Issue Date: ||11-Nov-2013|
|Series/Report no.: ||Nigerian Journal of Technological Research;Vol 1 no. 1 pp 62-66. 2002|
|Abstract: ||The study demonstrates the effect of ‘outliers’ on the accuracy of a model specified using the multiple regression analysis. Using the traffic volume from Minna to 27 other destinations, the model indicates that the mean trip rate, rail connectivity and bus trips per day account for 70% of the variations in the traffic volume, Further examination of the data reveals that Suleja contain unusually high daily bus trip rate of 12 and Lapai a high mean trip rate of 23.4. The two towns were dropped from the analysis and the explanatory power of the model increased appreciably to R2 = 83.60%. The study suggests the use of sufficiently large sample size, say a minimum of 50, as a way of eliminating the influence of extreme values in modeling.|
|Appears in Collections:||Urban and Regional Planning|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.