耶日·奈曼Jerzy Neyman

耶日·奈曼的父母是Czeslaw Neyman，他是一名律师，以及Kazimiera Lutoslawska。Kazimiera来自基辅，事实上，Czeslaw和Kazimiera是在他在基辅学习法律并租住Kazimiera母亲经营的房子里的一个房间时相识的。奈曼是他父母四个孩子中最小的，但由于最大的孩子Karol在奈曼出生时已经十六岁，而且两个女孩此时已经去世，他基本上是作为独生子被抚养长大的。值得注意的是，尽管当时存在政治局势（官方上波兰并不作为一个独立国家存在），奈曼出生在一个自认为是波兰人且肯定讲波兰语的天主教家庭。我们还应该注意到，奈曼在几年内以Splawa-奈曼的名义发表论文，但他在30岁时去掉了第一部分（这更像是一种贵族标志）。他名字的俄语版本是Yuri Czeslawovich，他在学生时代就以这个名字为人所知。

作为一个年轻男孩，奈曼住在几个不同的城镇：首先是本杰里，然后是赫尔松，然后是克里米亚的梅利托波尔，到他八岁时，全家住在同样在克里米亚的辛菲罗波尔。直到十岁，他都在家里由一位女家庭教师教导，然后他进入了当地的文理中学（Gymnasium）。值得注意的是，此时他已经能说五种语言：波兰语、乌克兰语、俄语、法语和德语。然而，1906年，他的父亲死于心脏病发作，他的母亲现在几乎没有钱来抚养儿子，搬到了她有亲戚的哈尔科夫。奈曼在哈尔科夫的中学表现出色，他决定，可能是因为他有一位杰出的数学老师，他将在大学学习数学。在完成学业和进入大学之间，他进行了一次穿越奥地利和意大利的欧洲火车旅行。他的母亲一定努力存钱才能负担得起送他进行这次旅行。

奈曼于1912年秋开始在哈尔科夫大学学习。他对物理学和数学都感兴趣，起初他最喜爱的是实际的物理实验。1914年第一次世界大战爆发时，许多学生离开去服兵役，但奈曼因视力测试不合格而留在大学。他很快意识到自己不具备成为一名有才华的实验者所必需的手工灵巧性，而当那位给他讲授函数论的教授建议他阅读昂利·勒贝格的论文Leçons sur l'intégration et la recherche des fonctions primitivesⓉ（《积分学教程及原函数研究》）时，他立刻被迷住了。他写了一篇关于昂利·勒贝格积分的论文，于1915年7月完成并提交参评金质奖章（后来获奖）。在1915-16学年，谢尔盖·伯恩斯坦给他讲授概率论；谢尔盖·伯恩斯坦对奈曼影响很大，并鼓励他阅读卡尔·皮尔逊的The Grammar of Science。多年后，在评论卡尔·皮尔逊著作的新版本时，他回忆起了它的影响（例如见[4]）：-

我们是一群年轻人，我们失去了对正统宗教的信仰，不是出于任何推理，而是因为我们的神职人员的愚蠢。[但]我们并未摆脱教条主义，实际上准备相信权威，只要它不是宗教权威。阅读《科学语法》……之所以令人震撼，是因为……它以毫不妥协的方式攻击了各种权威。……初读时，正是这一方面打动了我们。这可能意味着什么？我们不习惯任何科学著作中有这种语气。这部作品是一场骗局，作者是一个大骗子吗？但我们的老师伯恩斯坦推荐了这本书；我们必须再读一遍。

1917年9月，完成本科学习后，奈曼留在哈尔科夫大学为学术生涯做准备。他讲课并协助辅导课，并开始对统计思想产生兴趣。然而，战争的最后一年、俄国革命和内战彻底打乱了大学的学术生活。条件极为艰苦，毫不奇怪，奈曼的健康开始恶化。医生诊断出肺结核，并让他去南方休养。他于1919年前往克里米亚，在旅途中遇到了一位俄罗斯姑娘奥尔加·索洛多夫尼科娃。1919年秋他返回哈尔科夫开始教书后不久，奥尔加也回来了，他们的友谊促成了1920年5月4日的婚姻。然而，欧洲复杂的政治局势又出现了新的转折，在他们结婚时，波兰和俄国正在交战。结婚十天后，奈曼被监禁了大约六周。

尽管身处困境，奈曼还是通过了考试，成为哈尔科夫大学的讲师，教授高等代数、积分和集合论。由于担心自己即将再次被捕，他逃离了哈尔科夫，但在得知自己已经安全后又返回了。1921年，他前往波兰，并与瓦茨瓦夫·谢尔宾斯基取得联系，后者说那年晚些时候可能会有学术职位。为了挣钱，他在比得哥什（德语名Bromberg）的农业研究所担任高级统计助理。由于渴望到华沙，他于1922年12月在那里国家气象研究所找到了一份工作。他开始撰写关于统计应用的论文。1923-24学年初，他被任命为华沙大学的助理，同时还在农学院任教。1924年，他因一篇关于将probability应用于农业实验的学位论文而获得博士学位，答辩委员会包括瓦茨瓦夫·谢尔宾斯基和斯特凡·马祖尔凯维奇。

获得洛克菲勒奖学金前往伦敦与卡尔·皮尔逊合作后，他于1925年9月抵达那里，但失望地发现Pearson对现代数学一无所知。不过，他确实与伊根·皮尔逊——卡尔·皮尔逊的儿子——成为了朋友。

伊根·皮尔逊在3中写道，描述了此时的奈曼：-

我所记得的……是1926年春天，我们一起在萨里郡利思山上科尔德港的家族度假小屋（旧校舍）度过的那个周末。正是在那时，我着迷地听他讲述他在俄罗斯的早年生活，以及他后来在那些由战争和俄国革命在中欧引发的破坏性力量的阴影下所经历的一切。

奈曼对伦敦大学学院所学的统计学中缺乏数学感到失望，于是获得奖学金延期，以便在巴黎度过一年。他于1926年夏天抵达巴黎，拜访埃米尔·博雷尔。在巴黎的1926-27学年，奈曼听了埃米尔·博雷尔、昂利·勒贝格（他特别喜欢他的课）和雅克·阿达马的讲座，他的兴趣开始回到集合、测度和积分上。然而，伊根·皮尔逊的一封信再次激发了他对统计学的兴趣，伊根·皮尔逊寻求一个可以推导出戈塞特检验的一般原则。奈曼接着在假设检验方面产生了基础性成果，当伊根·皮尔逊于1927年春天访问巴黎时，他们合作撰写了第一篇论文。

奈曼于1927年5月回到波兰，立即试图在华沙建立一个生物统计实验室。他先后在华沙和克拉科夫度过时光，并于6月26日获得他的教授资格论文（Habilitation），开始作为讲师（Docent）讲课。1927年6月，奈曼写道：-

当然，[实验室]还不确定，尤其是我们在美国的贷款尚未签署。

然而到1928年，他已在华沙的嫩茨基实验生物学研究所建立了一个生物统计实验室。他与伊根·皮尔逊的合作继续，On the problem of two samples于1930年由克拉科夫的波兰科学院出版。然而波兰的生活变得越来越艰难，1931年奈曼写信给伊根·皮尔逊：-

……我们在波兰的一切都陷入了严重的危机。因此，政府通常拨给内恩茨基研究所的资金将大幅削减，我将难以养活我的‘小狗们’。

奈曼的‘小狗们’就是他实验室里的研究人员！到1932年，情况对奈曼来说似乎更糟了，他写道：——

我简直无法工作，危机和生存斗争占据了我所有的时间和精力。

1928年至1933年间，奈曼和伊根·皮尔逊撰写了许多关于假设检验的重要论文，这一合作成果丰硕，产生了诸如On the problem of the most efficient tests of statistical hypotheses（1933年）和The testing of statistical hypotheses in relation to probabilities a priori（1933年）等论文。

1933年，卡尔·皮尔逊从伦敦大学学院的弗朗西斯·高尔顿统计学讲席退休，伊根·皮尔逊成为应用统计系主任。奈曼于1934年获得三个月休假前往英国，填补伊根·皮尔逊系里的临时职位。次年该职位转为永久，奈曼一直担任至1938年。

将罗纳德·费希尔和伊根·皮尔逊安排在同一栋伦敦大学学院大楼里的做法，在个人层面上并不顺利。尽管罗纳德·费希尔曾启发奈曼的许多工作，但如今他们在同一栋楼里工作，关系似乎破裂了。1937年春，奈曼在美国进行了为期六周的大学巡回演讲，由塞缪尔·S·威尔克斯组织。回来后不久，他的一篇重要论文Outline of a theory of statistical estimation based on the classical theory of probability被皇家学会接受发表。他希望获得伦敦大学学院的讲席，但这似乎不太可能，因此他也考虑返回华沙。然而，伊根·皮尔逊努力将他留在伦敦，并设法安排了加薪。随后奈曼收到Evans的邀请，提供加州大学乔治·伯克利分校的讲师职位。

1970年，奈曼获得了荣誉学位，推荐信中列举了他在伦敦的成就（见[4]）：——

在1934至1938年间，奈曼对统计学做出了四项根本性贡献。每一项都足以确立国际声誉，无论是因为其直接影响，还是因为这些新思想和方法对年轻人和年长者思考的推动。他提出了置信区间理论，其在统计理论和数据分析中的重要性怎么强调都不为过。他对传染分布理论的贡献在解释生物学数据方面仍然非常有用。他关于分层总体抽样的论文为一种统计理论铺平了道路，这种理论除其他外，还为我们带来了盖洛普民意调查。[他的]工作，以及罗纳德·费希尔的工作，各自为随机实验提出了不同的模型，引领了在农业、生物学、医学和物理科学中广泛使用的全新实验领域。

1938年4月21日，奈曼接受了伯克利的聘请，并于8月抵达那里；此后他一直在伯克利工作。他在那里的数学系教授概率论与统计学，但目标是建立一个培养美国统计学家的中心。不久，数学系内出现了一个得到正式承认的统计实验室，但第二次世界大战的爆发使大多数学者偏离了原定方向。奈曼从事军事研究，特别是研究轰炸瞄准器和目标选择问题。战争结束时，他正要在伯克利重新着手建设统计学，却接到要求，作为观察希腊选举的小组成员前往希腊。显然，统计学家在这样的团队中会非常宝贵。

就奈曼而言，希腊之行并不很成功。他联系了一位希腊国际法教授，他知道这位教授在1926—27年他本人在巴黎时也在那里。由于这位教授在选举中并非中立，报纸上刊登了文章，而奈曼因不服从不得回应的命令而被解职。回到美国后，他重新努力在伯克利建立一所世界领先的数理统计学学派。他为实现目标不知疲倦地工作，统计学于1955年7月1日在伯克利成为一个独立的系。在[4]中引用了伯克利助理校长的一份备忘录：-

在这里，一位任性、执着且杰出的主任，在十五年间一步一步地，不顾其系主任和院长的意愿，把一个小小的“实验室”或研究所变成了就所教学生人数而言耗资巨大的单位；然后他又主张该单位应改名为“系”，因为不会产生额外费用。

奈曼在职业生涯后期对统计学研究的贡献，大多在气象学和医学应用领域。他因卓越贡献获得许多荣誉。我们特别注意到1966年的皇家统计学会盖伊奖章和1969年1月约翰逊总统颁发的科学奖章。

Jerzy Neyman's parents were Czeslaw Neyman, who was a lawyer, and Kazimiera Lutoslawska. Kazimiera came from Kiev and, in fact, Czeslaw and Kazimiera met when he was studying law at Kiev and rented a room in a house run by Kazimiera's mother. Jerzy was the youngest of his parents four children but, since the oldest child Karol was sixteen when Jerzy was born and two girls had by this time died, he was essentially brought up as an only child. It is worth noting that despite the political situation which existed at the time (officially Poland did not exist as a separate country), Jerzy was born into a Roman Catholic family which considered itself to be Polish and was certainly Polish speaking. We should also note that Neyman wrote papers for a few years under the name Splawa-Neyman, but he dropped the first part (which is more like a sign of nobility) at the age of 30. The Russian version of his name was Yuri Czeslawovich and he was known by this name when he was a student.

As a young boy, Jerzy lived in several different towns: first Bendery, then Kherson, then Melitopol in the Crimea, and by the time he was eight years old the family lived at Simferopol also in the Crimea. Up to the age of ten he was taught at home by a governess and then he entered the local gymnasium. Remarkably he could speak five languages by this time, Polish, Ukrainian, Russian, French and German. In 1906, however, his father died of a heart attack and his mother, now having little money to bring up her son, moved to Kharkov where she had relatives. Neyman excelled at the gymnasium at Kharkov and he decided, probably because he had an outstanding mathematics teacher, that he would study mathematics at university. Between completing his schooling and entering university, he made a European train journey through Austria and Italy. His mother must have saved hard to have been able to afford to send him on this trip.

Neyman began his studies at Kharkov University in the autumn if 1912. He was interested both in physics and in mathematics and at first it was practical physics experiments which he enjoyed most. Many students left for military service when World War I started in 1914 but Neyman failed the eyesight test so remained at University. He was quickly coming to realise that he did not have the necessary manual dexterity to be a talented experimenter and when the professor who was lecturing to him on the theory of functions suggested that he read Lebesgue's paper Leçons sur l'intégration et la recherche des fonctions primitives Ⓣ he immediately became fascinated. He wrote a paper on Lebesgue integration which he completed in July 1915 and submitted for the Gold Medal (which it later won). In the academic year 1915-16 Aleksandr Bernstein lectured to him on probability; he strongly influenced Neyman and encouraged him to read Karl Pearson's The Grammar of Science. Reviewing a new edition of Pearson's book many years later he recalled its impact (see for example [4]):-

We were a group of young men who had lost our belief in orthodox religion, not from any sort of reasoning, but because of the stupidity of our priests. [But] we were not freed from dogmatism and were prepared in fact to believe in authority, so far as it was not religious. The reading of The Grammar of Science ... was striking because ... it attacked in an uncompromising manner all sorts of authorities. ... At first reading it was this aspect that struck us. What could it mean? We were not used to this tone in any scientific book. Was the work a hoax and the author a scoundrel on a grand scale? But our teacher, Bernstein, had recommended the book; we must read it again.

In September 1917, having completed his undergraduate studies, Neyman remained at Kharkov University preparing for an academic career. He lectured and assisted with tutorial sessions and began to take an interest in statistical ideas. However the last year of the war, the Russian Revolution, and the civil war, totally disrupted the academic life of the University. There was great hardship and not surprisingly Neyman's health began to deteriorate. The doctors diagnosed tuberculosis and he was told to go south to recover. He travelled to the Crimea in 1919 and on the journey he met a Russian girl Olga Solodovnikova. Shortly after he returned to Kharkov to begin teaching in the autumn of 1919, Olga returned too and their friendship led to marriage on 4 May 1920. However the complex political situation in Europe had taken another twist and at the time of their marriage Poland and Russia were at war. Ten days after he was married, Neyman was imprisoned and held for about six weeks.

Despite the difficulties that he was under, Neyman passed his examinations and became a lecturer at Kharkov University, teaching higher algebra, integration, and set theory. Fearing he was about to be arrested again, he fled from Kharkov but returned on learning it was now safe for him. In 1921 he went to Poland and made contact with Sierpiński who said that an academic post later that year might be possible. To earn some money he took a job as senior statistical assistant at the Agricultural Institute in Bydgoszcz (German name Bromberg). Keen to be in Warsaw, in December 1922 he took a job in the State Meteorological Institute there. He began to write papers on applications of statistics. He was appointed as an assistant at Warsaw University at the beginning of the academic year 1923-24 and he also taught at the College of Agriculture. He received a doctorate in 1924 for a thesis on application of probability to agricultural experimentation after being examined by a panel which included Sierpiński and Mazurkiewicz.

Receiving a Rockefeller Fellowship to work with Karl Pearson in London, he arrived there in September 1925 but was disappointed to discover that Pearson was ignorant of modern mathematics. He did however become friendly with Egon Pearson, Karl Pearson's son.

Egon Pearson, writing in [3], describes Neyman at this time:-

What I remember ... is a week-end which we spent together in the spring of 1926 at our family holiday cottage (the Old School House) at Coldharbour on Leith Hill in Surrey. It was then that I listened with fascination to an account of his early life in Russia and of the experiences which he had later undergone in the shadow of those disruptive forces, set in train throughout Central Europe by war and the Russian Revolution.

Disappointed at the lack of mathematics in the statistics being studied at University College, London, Neyman obtained an extension of his fellowship to allow him to spend a year in Paris. He arrived in Paris in the summer of 1926 to visit Borel. In Paris for session 1926-27 Neyman attended lectures by Borel, Lebesgue (whose lectures he particularly enjoyed) and Hadamard and his interests began to move back towards sets, measure and integration. However his interest in statistics was stimulated again by a letter from Egon Pearson, who sought a general principle from which Gosset's tests could be derived. Neyman went on to produce fundamental results on hypothesis testing and, when Egon Pearson visited Paris in the spring of 1927, they collaborated in writing their first paper.

Neyman returned to Poland in May 1927 and immediately tried to set up a biometric laboratory in Warsaw. He spent time in both Warsaw and Kraków and on 26 June obtained his habilitation and began lecturing as a docent. In June 1927 Neyman wrote:-

Certainly [the laboratory] is not yet sure, especially as our loan in America is not yet signed.

However by 1928 he had managed to set up a Biometric Laboratory at the Nencki Institute for Experimental Biology in Warsaw. His collaboration with Egon Pearson continued with On the problem of two samples being published by the Polish Academy of Sciences in Kraków in 1930. However life in Poland was becoming increasingly hard and 1931 Neyman wrote to Egon Pearson:-

... we have in Poland a terrific crisis in everything. Accordingly the money from the Government given usually to the Nencki Institute will be diminished considerably and I shall have difficulties in feeding my pups.

Neyman's 'pups' were the research workers in his laboratory! By 1932 things seemed even worse for Neyman who wrote:-

I simply cannot work, the crisis and the struggle for existence takes all my time and energy.

Between 1928 and 1933 Neyman and Egon Pearson had written a number of important papers on hypothesis testing and the collaboration was highly productive with papers such as On the problem of the most efficient tests of statistical hypotheses (1933) and The testing of statistical hypotheses in relation to probabilities a priori (1933).

In 1933 Karl Pearson retired as from the Galton Chair of Statistics in University College London and Egon Pearson became Head of the Department of Applied Statistics. Neyman obtained a three month leave of absence to go in England in 1934 to fill a temporary post in Egon Pearson's department. The following year the post was made permanent and Neyman held it until 1938.

The arrangement which put Fisher and Egon Pearson into the same building at University College did not work well on a personal level. Although Fisher had inspired much of Neyman's work, now that they were working in the same building relations seemed to break down. In the spring of 1937 Neyman spent six weeks in the United States on a lecture tour of universities organised by Wilks. Shortly after his return he had a major paper Outline of a theory of statistical estimation based on the classical theory of probability accepted for publication by the Royal Society. He hoped to obtain a chair at University College but that seemed unlikely, so he also considered returning to Warsaw. However, Egon Pearson worked hard to keep him in London and managed to arrange salary increases. Then Neyman received an offer from Evans of a lectureship at the University of California at Berkeley.

In 1970 Neyman received an honorary degree and the letter of recommendation listed his achievements in London (see [4]):-

During the years 1934-38 Neyman made four fundamental contributions to the science of Statistics. Each of them would have been sufficient to establish an international reputation, both for their immediate effect and for the impetus which the new ideas and methods had on the thinking of young and old alike. He put forward the theory of confidence intervals, the importance of which in statistical theory and analysis of data cannot be overemphasised. His contribution to the theory of contagious distributions is still of great utility in the interpretation of biological data. His paper on sampling stratified populations paved the way for a statistical theory which, among other things, gave us the Gallup Poll. [His] work, and that of Fisher, each with a different model for randomised experiments, led to the whole new field of experimentation so much used in agriculture, biology, medicine, and physical sciences.

On 21 April 1938 Neyman accepted the offer from Berkeley and he arrived there in August; he worked in Berkeley for the rest of his life. There he taught probability and statistics in the mathematics department but aimed to set up a centre to train American statisticians. Soon there was an officially recognised Statistical Laboratory within the Department of Mathematics but the start of World War II diverted most academics from their intended course. Neyman undertook military research, particularly working on bomb sights and targeting problems. He was just getting back to building statistics at Berkeley at the end of the war when he received a request to go to Greece as part of a team to observe the elections there. Clearly a statistician would be invaluable on such a team.

The mission to Greece was not a great success as far as Neyman was concerned. He contacted a Greek professor of international law whom he knew to have been in Paris when he was there in 1926-27. Since the professor was not neutral in the elections, articles were published in the press, and Neyman was dismissed when he disobeyed orders not to respond. Back in the United States, he resumed his efforts to build a world-leading school of mathematical statistics at Berkeley. He worked tirelessly towards his goal and statistics became a separate department at Berkeley on 1 July 1955. In [4] a memo from the assistant president of Berkeley is quoted:-

Here, a wilful, persistent and distinguished director has succeeded, step by step over a fifteen year period, against the wish of his department chairman and dean, in converting a small 'laboratory' or institute into, in terms of numbers of students taught, an enormously expensive unit; and he then argues that the unit should be renamed a 'department' because no additional expense will be incurred.

Neyman's contributions to research in statistics over the latter part of his career were mostly in the areas of applications to meteorology and medicine. He received many honours for his remarkable contributions. In particular we note the Guy Medal of the Royal Statistical Society in 1966 and the Medal of Science from President Johnson in January 1969.