Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.561504
Title: Simultaneous confidence bands in linear regression analysis
Author: Ah-Kine, Pascal Soon Shien
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2010
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
A simultaneous confidence band provides useful information on the plausible range of an unknown regression model. For a simple linear regression model, the most frequently quoted bands in the statistical literature include the two-segment band, the three-segment band and the hyperbolic band, and for a multiple linear regression model, the most com- mon bands in the statistical literature include the hyperbolic band and the constant width band. The optimality criteria for confidence bands include the Average Width criterion considered by Gafarian (1964) and Naiman (1984) among others, and the Minimum Area Confidence Set (MACS) criterion of Liu and Hayter (2007). A concise review of the construction of two-sided simultaneous confidence bands in simple and multiple linear re- gressions and their comparison under the two mentioned optimality criteria is provided in the thesis. Two families of confidence bands, the inner-hyperbolic bands and the outerhyperbolic bands, which include the hyperbolic and three-segment bands as special cases, are introduced for a simple linear regression. Under the MACS criterion, the best con- fidence band within each family is found by numerical search and compared with the hyperbolic band, the best three-segment band and with each other. The inner-hyperbolic family of confidence bands, which include the hyperbolic and constant-width bands as special cases, is also constructed for a multiple linear regression model over an ellipsoidal covariate region and the best band within the family is found by numerical search. For a multiple linear regression model over a rectangular covariate region (i.e. the predictor variables are constrained in intervals), no method of constructing exact simultaneous con- fidence bands has been published so far. A method to construct exact two-sided hyperbolic and constant width bands over a rectangular covariate region and compare between them is provided in this thesis when there are up to three predictor variables. A simulation method similar to the ones used by Liu et al. (2005a) and Liu et al. (2005b) is also provided for the calculation of the average width and the minimum volume of confidence set when there are more than three predictor variables. The methods used in this thesis are illustrated with numerical examples and the Matlab programs used are available upon request.
Supervisor: Liu, Wei Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.561504  DOI: Not available
Keywords: QA Mathematics
Share: