这篇讣告回顾了丹麦理论统计学家Barndorff-Nielsen的生平与学术贡献，包括指数族理论、鞍点逼近、金融波动模型和ambit过程，适合对统计学史或统计理论感兴趣的读者。
Professor Ole Eiler Barndorff-Nielsen, who died on 26 June 2022, was arguably the most important Danish theoretical statistician in the 20th century. He was born on 18 March 1935 in Copenhagen. In 1956 he married his wife Bente, with whom he had three children, and whose persistent and wholehearted support was an essential prerequisite for his illustrious scientific career. Barndorff-Nielsen became interested in statistics when, as a student of actuarial mathematics at the University of Copenhagen, he worked part-time for Michael Weis Bentzon at the Department of Biostatistics of the Danish State Serum Institute (1954-1958). In this period, he also met and was influenced by Georg Rasch. He graduated in 1960 with a degree in mathematics and mathematical statistics from the University of Aarhus, where he spent most of his academic life, and where he became professor of statistics in 1973. He spent extended periods at the University of Minnesota, Stanford University, University of Cambridge, and the Technical University of Munich. Barndorff-Nielsen's role in the development of statistics in Denmark cannot be overestimated. A main achievement was the creation of a strong statistics environment at the University of Aarhus. No previous teaching of theoretical statistics existed and courses had to be developed from scratch. Simultaneously, a group of pioneering young researchers was trained to become the first statistics group at the Department of Mathematics in Aarhus. For several years, a high-level curriculum for theoretical statistics was developed in close collaboration with colleagues at the University of Copenhagen, where a Department of Mathematical Statistics had been created in 1960. A specific achievement was the internationalisation of Danish statistics, in which Barndorff-Nielsen played a leading and decisive role. During his entire career he collaborated with researchers in many parts of the world and ensured that his younger colleagues had the possibility to do the same. Over the years he gave invited lectures at numerous universities and conferences, including the Forum lectures at the European Meeting of Statisticians in 1981. He had a particularly strong connection to the British statistics environment, and was an Honorary Fellow of the Royal Statistical Society. He organised or co-organised a long list of international scientific meetings and was instrumental in the development of a number of international scientific organisations. He was one of the founders of the Nordic Conferences in Mathematical Statistics, a series of biennial conferences that is still vibrant in their tradition. The first conference was held in Aarhus in 1965. Later he became a leading force in the creation and development of the Bernoulli Society. While he was chairman of the European Regional Committee in the late 1970s, he wrote the scenario that defined the form and type of the European Meetings of Statisticians; a scenario that, with a few minor updates, is still in use. In the ISI, Barndorff-Nielsen was a main force behind turning the International Statistical Review into a highly respected research journal, of which he was the editor in the years 1980–1987. In 1993–1995 he was a very influential president of the Bernoulli Society, where he strengthened the organisation in many ways. Among many other initiatives, he started the Bernoulli journal and the Newsletter of the Bernoulli Society. He was the first editor of Bernoulli in 1994–2000 and ensured the continuing success of the journal. Among Barndorff-Nielsen's early scientific contributions is his work on the exact theory of exponential families and on the foundations of statistics, in particular sufficiency and conditional inference. His book Information and Exponential families in Statistical Theory, published by Wiley in 1978, is still the main reference on exponential families. Being the first international research monograph originating from Denmark, the book became a milestone that marks the beginning of the internationalisation that has since been a hallmark of statistical research in Denmark. In 1977 he introduced the hyperbolic distribution as a mathematical model of the size-distribution of sand grains. This distribution formalised heuristic ideas by Brigadier Ralph Alger Bagnold, but in a way characteristic of his research, Barndorff-Nielsen presented it as a member of the class of generalised hyperbolic distributions, defined as normal variance-mean mixtures. These distributions, in particular the normal-inverse Gaussian distribution, have later turned out to be widely applicable in many other fields, in particular turbulence and finance. Later, Barndorff-Nielsen played a significant role in developing the asymptotic theory of statistics. A main contribution is his work on saddle point approximation methods in statistics, not least his formula for the conditional distribution of the maximum likelihood estimator given an (approximately) ancillary statistic that generalises a formula by R.A. Fisher. It was originally called the p*-formula, but is now widely known as Barndorff-Nielsen's formula. Jointly with Sir David Cox he wrote two influential books on asymptotic techniques in statistics. Relatively late in his career, Barndorff-Nielsen started to work on stochastic models in finance. A main contribution was the Barndorff-Nielsen–Shephard model for the price of a financial asset, where the random volatility was modelled by an Ornstein-Uhlenbeck process driven by a Lévy-process or a sum of such processes. Jointly with Neil Shephard and other collaborators he started and contributed very significantly to the development of statistical methods for high frequency financial time series. Barndorff-Nielsen's contributions to financial econometrics were presented in the joint book with Neil Shephard Financial Volatility in Continuous Time published in 2012. In this phase of his life, Barndorff-Nielsen also returned with great energy and originality to his previous interest in turbulence. His activities in this area ranged from analysis of turbulence data and collaboration with the company Siemens Wind Power to the development of a new model type that he called ambit-processes. These are processes, where the present state is given as an integral of a so-called Lévy basis over a set, the ambit set, of points in space-time that are able to influence the present state. In recent years Barndorff-Nielsen devoted most of his energy to the development of a stochastic calculus for ambit processes. Characteristically, he inspired a group of talented young researchers to work with him on what has developed into a vibrant area of research. Results of these efforts were presented in the joint book with Fred Espen Benth and Almut Veraart entitled Ambit Stochastics and published by Springer in 2018. Barndorff-Nielsen received a large number of national and international honours. He was a member of the Royal Danish Academy of Sciences and Letters and of Academia Europaea, and a Corresponding Fellow of the Royal Society of Edinburgh. He received honorary doctoral degrees from Université Paul Sabatier in Toulouse and Katholieke Universiteit Leuven. In 2001 he received the German Humboldt-prize and in 2010 the Faculty Prize from the Faculty of Science, Aarhus University. He was an Honorary Fellow of the Royal Statistical Society, a Fellow of the Institute of Mathematical Statistics, an elected member of the International Statistical Institute, and an Honorary Member of the Danish Society for Theoretical Statistics. The size and range of Barndorff-Nielsen's contributions to research and science organisation are of awe-inspiring proportions. That he was able to accomplish all this was due to his diligence, persistence and strong personality. He had a natural authority that made him the unquestioned leader in any context. Anyone who worked with him quickly learned how difficult it was to say no to him. He worked simultaneously on a large number of problems and projects, which was possible because he was extremely well-organised. Visitors to his office were impressed by the many large, but meticulous, stacks of paper on his desk, from which he was able to quickly retrieve exactly the neat paper-folder he was looking for. Barndorff-Nielsen used his influence and authority to strengthen the statistical research community both in Denmark and worldwide. He was, not least, tireless in his support of the next generations of statisticians. He saw this as one of his most important responsibilities. A large number of statisticians, Danish as well as from other countries, who at a young age met Barndorff-Nielsen and worked with him, remember fondly his caring support at many levels. The open, inquisitive and hard working environment he created around himself, made a lasting impression on anyone who had the privilege of working with him.