Sunday 27 November 2016

Book review: The Drunkard's Walk

I just finished reading the 2008 book by Leonard Mlodinow "The Drunkard's Walk - How Randomness Rules Our Lives". I was kind of expecting something like the Nassim Nicholas Taleb book, "Fooled by Randomness" (which I reviewed earlier this year), but this book was much better.

While it purports to be a book about the role of randomness in our lives, much of the content is a wide-ranging and well-written series of stories in the history of probability theory and statistics. To pick out the highlights (to me), Mlodinow covers the Monty Hall problemPascal's triangle, Bernoulli's 'golden theorem' (better known as the law of large numbers), Bayes' theorem, the normal distribution and the central limit theorem, regression to the mean and the coefficient of correlation (both developments by Francis Galton, known as the father of eugenics), Chi-squared tests (from Karl Pearson), Brownian motion, and the butterfly effect. And all of that in a way that is much more accessible than the Wikipedia links I've provided.

Throughout the book, Mlodinow illustrates his points with interesting anecdotes and links to relevant research. One section, on measurement issues, struck me in particular because it made me recall one of the funniest papers I have ever read, entitled "Can People Distinguish Pâté from Dog Food?" (the answer was no). Mlodinow focuses on the subjectivity (and associated randomness) of wine ratings:
Given all these reasons for skepticism, scientists designed ways to measure wine experts' taste discrimination directly. One method is to use a wine triangle. It is not a physical triangle but a metaphor: each expert is given three wines, two of which are identical. The mission: to choose the odd sample. In a 1990 study, the experts identified the odd sample only two-thirds of the time, which means that in 1 out of 3 taste challenges these wine gurus couldn't distinguish a pinot noir with, say, "an exuberant nose of wild strawberry, luscious blackberry, and raspberry," from one with "the scent of distinctive dried plums, yellow cherries, and silky cassis." In the same study an ensemble of experts was asked to rank a series of wines based on 12 components, such as alcohol content, the presence of tannins, sweetness, and fruitiness. The experts disagreed significantly on 9 of the 12 components. Finally, when asked to match wines with the descriptions provided by other experts, the subjects were correct only 70 percent of the time.
Clearly, randomness is at play more often than we probably care to admit. So, apparently at random, I highly recommend this book (and I've made sure to add some of Mlodinow's other books to my Amazon wish list to pick up later!).

No comments:

Post a Comment