Abstract
IN a paper read before the Royal Statistical Society on December 18, Prof. R. A. Fisher surveyed the recent change in the outlook of mathematical statisticians. The most profound modification seems to have taken place rather in the logical than in the mathematical aspect, though it has been brought about by the resolution of mathematical difficulties. Statisticians are now dealing with types of uncertain inference wider than those of the theory of probability. Prof. Fisher expressed the view that the current teaching of pure mathematics is not an altogether adequate preparation for fruitful work in this field, for this teaching is purely deductive, omitting the essential concepts of inductive logic, and insists on ‘rigour’ in a limited sense which he considers very inadequate to the requirements of an inductive problem. The questions raised by Prof. Fisher are of great interest and importance, but it should not be overlooked that there is still some difference of opinion concerning them, as will be seen by a perusal of the series of papers in the Proceedings of the Royal Society (1932-4) by Dr. H. Jeffreys.
Article PDF
Rights and permissions
About this article
Cite this article
Statistics and Inductive Inference. Nature 135, 61 (1935). https://doi.org/10.1038/135061a0
Issue Date:
DOI: https://doi.org/10.1038/135061a0