Vanishing shortcoming of data driven Neyman's tests

Tadeusz Inglot; B. Szyszkowicz; W.C.M. Kallenberg; Teresa Ledwina

Vanishing shortcoming of data driven Neyman's tests

Tadeusz Inglot, B. Szyszkowicz (Editor), W.C.M. Kallenberg, Teresa Ledwina

Research output: Contribution to conference › Paper › peer-review

Abstract

The shortcoming of a test is the difference between the power of the test and the power of the most powerful test. For a large set of alternatives converging to the null hypothesis asymptotic optimality of data driven Neyman's tests is shown in terms of vanishing shortcoming when the level of signicance tends to zero. In contrast to classical goodness-of-fit tests data driven Neyman's tests are asymptotically efficient in an infinite number of orthogonal directions.

Original language	Undefined
Pages	811-829
Number of pages	19
Publication status	Published - 1998
Event	Asymptotic Methods in Probability and Statistics - Ottawa Duration: 8 Jul 1997 → 13 Jul 1997

Conference

Conference	Asymptotic Methods in Probability and Statistics
Period	8/07/97 → 13/07/97
Other	8-13 July 1997

Keywords

intermediate eciency
MSC-62G10
IR-62414
Shortcoming
Schwarz's criterion
Large deviations
MSC-62G20
smooth test
EWI-13167

Access to Document

http://eprints.eemcs.utwente.nl/secure2/13167/01/vanishing_shortcoming_of_data_driven_Neymans_tests_memo1393.pdf

Cite this

@conference{2295b6b21bec4bc884f3f23dc5ed7db2,

title = "Vanishing shortcoming of data driven Neyman's tests",

abstract = "The shortcoming of a test is the difference between the power of the test and the power of the most powerful test. For a large set of alternatives converging to the null hypothesis asymptotic optimality of data driven Neyman's tests is shown in terms of vanishing shortcoming when the level of signicance tends to zero. In contrast to classical goodness-of-fit tests data driven Neyman's tests are asymptotically efficient in an infinite number of orthogonal directions.",

keywords = "intermediate eciency, MSC-62G10, IR-62414, Shortcoming, Schwarz's criterion, Large deviations, MSC-62G20, smooth test, EWI-13167",

author = "Tadeusz Inglot and B. Szyszkowicz and W.C.M. Kallenberg and Teresa Ledwina",

year = "1998",

language = "Undefined",

pages = "811--829",

note = "null ; Conference date: 08-07-1997 Through 13-07-1997",

}

TY - CONF

T1 - Vanishing shortcoming of data driven Neyman's tests

AU - Inglot, Tadeusz

AU - Kallenberg, W.C.M.

AU - Ledwina, Teresa

A2 - Szyszkowicz, B.

PY - 1998

Y1 - 1998

N2 - The shortcoming of a test is the difference between the power of the test and the power of the most powerful test. For a large set of alternatives converging to the null hypothesis asymptotic optimality of data driven Neyman's tests is shown in terms of vanishing shortcoming when the level of signicance tends to zero. In contrast to classical goodness-of-fit tests data driven Neyman's tests are asymptotically efficient in an infinite number of orthogonal directions.

AB - The shortcoming of a test is the difference between the power of the test and the power of the most powerful test. For a large set of alternatives converging to the null hypothesis asymptotic optimality of data driven Neyman's tests is shown in terms of vanishing shortcoming when the level of signicance tends to zero. In contrast to classical goodness-of-fit tests data driven Neyman's tests are asymptotically efficient in an infinite number of orthogonal directions.

KW - intermediate eciency

KW - MSC-62G10

KW - IR-62414

KW - Shortcoming

KW - Schwarz's criterion

KW - Large deviations

KW - MSC-62G20

KW - smooth test

KW - EWI-13167

M3 - Paper

SP - 811

EP - 829

Y2 - 8 July 1997 through 13 July 1997

ER -