Anmerkungen:
Includes index. - Description based on print version record and CIP data provided by publisher; resource not viewed
Beschreibung:
"Prediction Revisited is a ground-breaking book for financial analysts and researchers--as well as data scientists in other disciplines--to reconsider classical statistics and approaches to forming predictions. Czasonis, Kritzman, and Turkington lay out the foundations of their cutting-edge approach to observing information from data. And then characterize patterns between multiple attributes, soon introducing the key concept of relevance. They then show how to use relevance to form predictions, discussing how to measure confidence in predictions by considering the tradeoff between relevance and noise. Prediction Revisited applies this new perspective to evaluate the efficacy of prediction models across many fields and preview the extension of the authors' new statistical approach to machine learning. Along the way they provide colorful biographical sketches of some of the key scientists throughout history who established the theoretical foundation that underpins the authors' notion of relevance--and its importance to prediction. In each chapter, material is presented conceptually, leaning heavily on intuition, and highlighting the key takeaways reframe prediction conceptually. They back it up mathematically and introduce an empirical application of the key concepts to understand. (If you are strongly disinclined toward mathematics, you can pass by the math and concentrate only on the prose, which is sufficient to convey the key concepts of this book.) In fact, you can think of this book as two books: one written in the language of poets and one written in the language of mathematics. Some readers may view the book's key insight about relevance skeptically, because it calls into question notions about statistical analysis that are deeply entrenched in beliefs from earlier training. The authors welcome a groundswell of debate and advancement of thought about prediction."--