Showing 1–6 of 6 results for author: Dergachev, V

Efficient representations of upper limits and confidence intervals for experimental data

Authors: Vladimir Dergachev

Abstract: Upper limits and confidence intervals are a convenient way to present experimental results. With modern experiments producing more and more data, it is often necessary to reduce the volume of the results for convenient distribution. A common approach is to take a maximum over a set of upper limits, which yields an upper limit valid for the entire set. This, however, can result in significant sensi… ▽ More Upper limits and confidence intervals are a convenient way to present experimental results. With modern experiments producing more and more data, it is often necessary to reduce the volume of the results for convenient distribution. A common approach is to take a maximum over a set of upper limits, which yields an upper limit valid for the entire set. This, however, can result in significant sensitivity loss, because we discard more constraining data. In this paper we introduce functional upper limits and confidence intervals that allow to summarize results with small relaxation of constraints. A toy example and an application to upper limits in all-sky continuous gravitational wave searches are worked out. △ Less

Submitted 7 September, 2024; v1 submitted 16 November, 2023; originally announced November 2023.

Comments: Expanded version with a toy example. Author preferred equation formatting

MSC Class: 62G15 (Primary); 49M99 (Secondary)

A Novel Universal Statistic for Computing Upper Limits in Ill-behaved Background

Authors: Vladimir Dergachev

Abstract: Analysis of experimental data must sometimes deal with abrupt changes in the distribution of measured values. Setting upper limits on signals usually involves a veto procedure that excludes data not described by an assumed statistical model. We show how to implement statistical estimates of physical quantities (such as upper limits) that are valid without assuming a particular family of statistica… ▽ More Analysis of experimental data must sometimes deal with abrupt changes in the distribution of measured values. Setting upper limits on signals usually involves a veto procedure that excludes data not described by an assumed statistical model. We show how to implement statistical estimates of physical quantities (such as upper limits) that are valid without assuming a particular family of statistical distributions, while still providing close to optimal values when the data is from an expected distribution (such as Gaussian or exponential). This new technique can compute statistically sound results in the presence of severe non-Gaussian noise, relaxes assumptions on distribution stationarity and is especially useful in automated analysis of large datasets, where computational speed is important. △ Less

Submitted 6 March, 2013; v1 submitted 9 August, 2012; originally announced August 2012.

Comments: 11 pages; expanded version of the original article

Report number: LIGO-P1200065-v8 MSC Class: 62-07; 62P35; 62G15; 62G32

A non radical based approach to study of associative algebras

Authors: Vladimir Dergachev

Abstract: We study pairs of associative algebras and linear functionals. New results together with corrected proofs for previously published material are presented. In particular, we prove the identity ind Mat(n) (#) A=n * ind A$ for finite-dimensional unital associative algebra A with index 1. [ Symbol (#) denotes tensor product ] We study pairs of associative algebras and linear functionals. New results together with corrected proofs for previously published material are presented. In particular, we prove the identity ind Mat(n) (#) A=n * ind A$ for finite-dimensional unital associative algebra A with index 1. [ Symbol (#) denotes tensor product ] △ Less

Submitted 18 April, 2003; v1 submitted 29 December, 2002; originally announced December 2002.

Comments: 40 pages; text has been significantly expanded. Added new sections discussing interaction with radical methods and open problems

MSC Class: 16B99

On a new approach to classification of associative algebras

Authors: Vladimir Dergachev

Abstract: In this paper we explore a new method of analysis of associative algebras. In this paper we explore a new method of analysis of associative algebras. △ Less

Submitted 2 March, 2002; originally announced March 2002.

Comments: 14 pages, Latex 2e

MSC Class: 17B99 (Primary)

On associative algebras with unity and Lie index 1

Authors: Vladimir Dergachev

Abstract: We present a new method of analysis of associative algebras. This method bears a certain resemblance to the famous analysis of commutative $C^*$-algebras in which an important role is played by multiplicative functionals over the algebra. These functionals can be considered as "rank 1" in the framework of our method. In our approach a special position is taken by generic (maximal rank) functiona… ▽ More We present a new method of analysis of associative algebras. This method bears a certain resemblance to the famous analysis of commutative $C^*$-algebras in which an important role is played by multiplicative functionals over the algebra. These functionals can be considered as "rank 1" in the framework of our method. In our approach a special position is taken by generic (maximal rank) functionals, which reveal a wealth of information on the structure of non-commutative algebras. We apply this method to derive interesting properties of index of Lie algebras and to analyze finite-dimensional associative algebras with unity and Lie index 1. △ Less

Submitted 1 August, 2001; v1 submitted 16 May, 2000; originally announced May 2000.

Comments: Removed faulty proof of $V(1)=Stab_F$ at the very end of section 5

MSC Class: 16B99; 17B99

Some properties of index of Lie algebras

Authors: Vladimir Dergachev

Abstract: In this paper we prove several theorems about the behavior of index of Lie algebras derived from associative algebras under tensor products of underlying associative algebras. In this paper we prove several theorems about the behavior of index of Lie algebras derived from associative algebras under tensor products of underlying associative algebras. △ Less

Submitted 7 January, 2000; originally announced January 2000.

Comments: 16 pages, Latex 2e

MSC Class: 17B99 (Primary) 16G99 (Secondary)