The Statistical Work of Oskar Anderson

Abstract

T HE death of Professor Oskar Anderson in Munich, Germany on February 12, 1960 in his 73rd year is a great loss for statisticians everywhere. Professor Anderson was born on August 2, 1887 in Minsk, Russia. He studied mathematics, physics, economics, and law at the Universities of Kasan and St. Petersburg (Leningrad). He was an assistant of the well known Russtian statistician A. A. Tschuprow. He taught mathematical statistics at the High School of Commerce in Kasan, later at the High School of Commerce in Varna (Bulgaria), the University of Kiel, and the University of Munich. Apart from his teaching activities he participated actively in sampling censuses in Russia and Bulgaria. He was perhaps the most widely known statistician in Central Europe. During his very distinguished career, he provided a link between the Russian school of statistics, from which he originated (Markoff, Tschuprow) and the Anglo-American school. He was an honorary doctor of the University of Vienna, also of the High School of Economics, Mannheim (Germany), an honorary member of the Royal Statistical Society, and of the German Statistical Society, fellow and founder of the Econometric Society, member of the International Statistical Institute, fellow of the American Statistical Society, the Institute of Mathematical Statistics, and the American Society for the Advancement of Science. Through his origin in the flourishing Russian school of probabilistes, who are certainly today the most important contributors in this field (Kolmogoroff, Khintchin, Gnedenko, etc.), Anderson belongs to the so-called continental school of statistics, and worked in the tradition of the well known German statisticians Lexis and von Bortkiewich. He might be the last representative of this approach, since recent textbooks in statistics published in German show already the overwhelming influence of the Anglo-American school of Sir Ronald Fisher and Jerzy Neyman. It is, of course, impossible to isolate schools of statistics in this manner, as actually the members of the various tendencies in statistics were in contact with each other. Anderson himself appreciated the achievements of the AngloAmerican school to a certain extent, and was also somewhat critical about the applicability of these methods in the field of the social sciences. We propose to discuss the works of Anderson under seven headings: (1) biographical and bibliographical, (2) probability, (3) survey sampling, (4) Variate Difference Method, (5) time series analysis, (6) econometrics, (7) index numbers, (8) correlation methods, (9) textbooks.