In his book The Signal and the Noise, Nate Silver explains the difference between good and bad statistical modelling. Described by the New York Times as one of the most momentous books of the decade, when published in 2012, it has now been updated for the 2020 pandemic and the most recent US presidential elections.
In the book, the author investigates how to distinguish a true signal from a myriad of noisy data and gives reasons why occasionally more data, if not analysed properly, can lead to worse predictions. And while the weather forecast is a success story, improving enormously in the past 25 years, some events like earthquakes and economic markets are almost bullet-proof to predictions. Nate Silver gives us strong reasons though, why we should never give up on statistical modelling. An enriching read for every person aiming to improve their statistical literacy.
