Strong confidence intervals for autoregression - Research - Royal Holloway, University of London

Home > Computer Science home > Research > Profile > Publications > Strong confidence intervals for autoregression

Strong confidence intervals for autoregression. / Vovk, Vladimir.

2007.

Research output: Working paper

Published

Standard

Strong confidence intervals for autoregression. / Vovk, Vladimir.

2007.

Research output: Working paper

BibTeX

@techreport{52ed02fd710b4000a404bc77697836ed,

title = "Strong confidence intervals for autoregression",

abstract = "In this short note I apply the methodology of game-theoretic probability to calculating non-asymptotic confidence intervals for the coefficient of a simple first order scalar autoregressive model. The most distinctive feature of the proposed procedure is that with high probability it produces confidence intervals that always cover the true parameter value when applied sequentially.",

keywords = "math.ST, stat.ME, stat.TH",

author = "Vladimir Vovk",

note = "7 pages, 2 tables, 2 figures",

year = "2007",

month = jul,

day = "4",

language = "English",

type = "WorkingPaper",

}

RIS

TY - UNPB

T1 - Strong confidence intervals for autoregression

AU - Vovk, Vladimir

N1 - 7 pages, 2 tables, 2 figures

PY - 2007/7/4

Y1 - 2007/7/4

N2 - In this short note I apply the methodology of game-theoretic probability to calculating non-asymptotic confidence intervals for the coefficient of a simple first order scalar autoregressive model. The most distinctive feature of the proposed procedure is that with high probability it produces confidence intervals that always cover the true parameter value when applied sequentially.

AB - In this short note I apply the methodology of game-theoretic probability to calculating non-asymptotic confidence intervals for the coefficient of a simple first order scalar autoregressive model. The most distinctive feature of the proposed procedure is that with high probability it produces confidence intervals that always cover the true parameter value when applied sequentially.

KW - math.ST

KW - stat.ME

KW - stat.TH

M3 - Working paper

BT - Strong confidence intervals for autoregression

ER -