A Comparison of Two Proximity Matrices

Abstract

本文的目的旨在提出一種資料分析策略的一般性理論架構稱作二次式分配準則(quadratic assignment paradigm,簡寫為QAP)，以考驗兩個方形近似矩陣(proximity matrices)之間的相似性。本文同時提出如何運用QAP技術的詳細計算過程與兩個應用實例，對於QAP技術的應用範圍亦加以扼要地評論。

The purpose of this paper is to present a general framework of data-analytic strategies, referred to as the quadratic assignment paradigm (QAP), that can be used to test a hypothesis that two or more proximity matrices on the same set of objects reflecting a similar patterning of high and low entries. The detailed computation procedure for executing the QAP technique is provided. Two empirical examples showing how the QAP technique is applied are given. Comments on the applications of the QAP technique are also summarized.\n