for History of Probability and Statistics

It is generally held that R.A. Fisher, Jerzy Neyman and Egon Pearson laid the foundations of statistical inference in the 1920s and 1930s. Recent research concerning the history of statistical ideas has discovered that already at the end of the 18th century Laplace sketched a method, called the principle of inverse probability, which involved characteristics features of statistical inference. Laplace’s estimation of the size of population in France, applying inverse probability principle, is the first scientifically ambitious partial investigation, or sample survey. Laplace’s investigation embraced many of the phases that modern sample surveys have. The inverse probability principle and his later contributions, such as the Central Limit Theorem, eventually laid foundations for the later development of statistical inference. Laplacian paradigm was the dominant model in statistical science up to the 1920s but since then it has fallen into oblivion – apparently, because it has been regarded as ‘Bayesian’. Recent research has shown that Laplace’s and Fisher’s ideas actually were close to each other even though they were based on different inference model.