On the distribution of the maximum of Brownian bridges with application to regression with correlated errors. (English) Zbl 0726.62151
Summary: A Brownian bridge of order q is the weak limit of a residual partial sum obtained from regression fitting. When \(q=0\), the process is the usual Brownian bridge, and the distribution of the maximum is known analytically. For \(q\geq 1\), the supremum distributions are approximated by a Monte Carlo technique. The theoretical results are then applied in two examples involving polynomial regression. Tables are given that allow a determination of the degree of regression fitting in the case of polynomial regression with autoregressive errors. This procedure is also a graphical procedure.
MSC:
62M10 | Time series, auto-correlation, regression, etc. in statistics (GARCH) |
65C05 | Monte Carlo methods |
60G15 | Gaussian processes |
62M99 | Inference from stochastic processes |
References:
[1] | Andrews D.A, A collection of problems from many fields for the student and research worker (1985) · Zbl 0567.62002 |
[2] | Billingsley, P. 1968. ”Convergence of probability measures”. New York: Wiley. · Zbl 0172.21201 |
[3] | Box G.E.P, Forcasting and control (1976) |
[4] | DOI: 10.1214/aoms/1177706645 · Zbl 0085.13720 · doi:10.1214/aoms/1177706645 |
[5] | Jandhyala V.K, Ph.D. Thesis (1985) |
[6] | DOI: 10.1016/0304-4149(85)90380-1 · Zbl 0584.62144 · doi:10.1016/0304-4149(85)90380-1 |
[7] | DOI: 10.2307/3214098 · Zbl 0633.62088 · doi:10.2307/3214098 |
[8] | DOI: 10.1080/03610928808829880 · Zbl 0696.62053 · doi:10.1080/03610928808829880 |
[9] | DOI: 10.1214/aos/1176344133 · Zbl 0375.62064 · doi:10.1214/aos/1176344133 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.