A NONPARAMETRIC TEST OF TREND BASED ON AMALGAMATION
Taka-aki Shiraishi, Akio Kudô · Memoirs of the Faculty of Science Kyushu University Series A Mathematics · 1981
In this paper we deal with a nonparametric test of trend proposed by Brunk [3]. This test is identical with the test based on the number of distinct values obtained by applying the Pool Adjacent Violators Algorithm or amalgamation process to the data. (c.f. Barlow et al. [2]) We provide rather extensive table, and an example is taken from Barlow et al. [2]. We conducted simulations and tables of estimated power is also provided.