Scholion.ai

Selection of Largest Multiple Correlation Coefficients: Asymptotic Case

M. Haseeb Rizvi, Herbert Solomon

Journal of the American Statistical Association · 1973 · Volume 68 · Issue 341

DOI: 10.1080/01621459.1973.10481360

Abstract

Selection of t largest from among k independent multiple correlation coefficients, , is provided by a procedure with the property that the probability of a correct selection is not smaller than a preassigned P * whenever the t largest θ i (1) exceed each remaining θ i by a magnitude δ 1 , and, simultaneously, (2) each θ i is at least δ 2 times as large as each remaining θ i . The separation thresholds, 0 < δ 1 < 1 and δ 2 > 1 are also preassigned. Several applications are discussed and tables are listed.

Enable JavaScript for citation tools and additional paper interactions.