Towards Predictive Statistics: A Pedagogical Explanation

Abstract

In statistics application area, lately, several publications appeared that warn about the dangers of the inappropriate application of statistics and remind the users of the recall that prediction is the ultimate objective of the statistical analysis. This trend is known as predictive statistics. However, while the intended message is aimed at the very general audience of practitioners and researchers who apply statistics, many of these papers are not easy to read since they are either too technical and/or too philosophical for the general reader. In this short paper, we describe the main ideas and recommendation of predictive statistics in – hopefully – clear terms. 1 Limitations of the Traditional Statistics and Need for Predictive Approach Prediction is important. One of the main objectives of science and engineering is to predict future events – i.e., to predict what will happen in general, and, specifically, predict what will happen if we undertake a certain action. Prediction in science and engineering. From this viewpoint, the progress of science and engineering is usually made as follows: • we analyze the existing data, and come up with formulas connecting different quantities, • we then use these formulas to predict new phenomena and/or future values of different quantities. If the prediction is successful, i.e., if the observed future values are indeed close to the predictions, then the theory is confirmed. This is how Mendeleev’s periodic table became an accepted theory: when Mendeleev used the observed periodicity to predict several new elements, and