On Two New Classes of Subset Selection Procedures for Location Parameters

Abstract

Kumar, Mehta, and Kumar (2002) developed two new classes of subset selection procedures for location parameters. In this paper, Pitman asymptotic relative efficiencies are computed for some members of these classes when compared among themselves and to the existing procedures, with interesting results. Approximate implementation of these procedures with the help of existing tables is illustrated. Simulation study is carried out to assess the sample size sufficient for their implementation.

Two New Classes of Subset Selection Procedures for Location Parameters

“…The statistic U"'' c can also be written in terms of ranks as follows: of Kumar et al [16]) and the constant b 'is chosen such that asymptotically…”

“…The book is a nice compendium covering a lot of the relevant material for introductory and advanced courses in time series analysis and contains a huge number of relevant references in every of the 16 A useful reference for this topic not cited by the author is the book "Sample-Size Determination" by E. A. Mace, Reinhold Publishing Corporation, New York, 1964. The next chapter deals with covariation with emphasis on regression and correlation.…”

“…For other references one may refer to Gupta and Panchapakesan [8], Dudewicz and Koo [5], Dudewicz [4] and Gupta and Panchapakesan [9]. [16].…”