Preface

The statistics profession is at a unique point in history. The need for valid statistical tools is greater than ever; data sets are massive, often measuring hundreds of thousands of measurements for a single subject. The field is ready for a revolution, one driven by clear, objective benchmarks under which tools can be evaluated. Statisticians must be ready to take on this challenge. They have to be dynamic and thoroughly trained in statistical concepts. More than ever, statisticians need to work effectively in interdisciplinary teams and understand the immense importance of objective benchmarks to evaluate statistical tools developed to learn from data. They have to produce energetic leaders who stick to a thorough a priori road map, and who also break with current practice when necessary. Why do we need a revolution? Can we not keep doing what we have been doing? Sadly, nearly all data analyses are based on the application of so-called parametric (or other restrictive) statistical models that assume the data-generating distributions have specific forms. Many agree that these statistical models are wrong. That is, everybody knows that linear or logistic regression in parametric statistical models and Cox proportional hazards models are specified incorrectly. In the early 1900s, when R.A. Fisher developed maximum likelihood estimation, these parametric statistical models were suitable since the data structures were very low dimensional. Therefore, saturated parametric statistical models could be applied. However, today statisticians still use these models to draw conclusions in high-dimensional data and then hope these conclusions are not too wrong.