if then [reading a book self-review]

Nature of 17 September 2020 has a somewhat surprising comment section where an author, Jill Lepore from Harvard University, actually summarises her own book, If Then: How the Simulmatics Corporation invented the Future. This book is the (hi)story of a precursor of Big Data Analytics, Simulmatics, which used as early as 1959 clustering and simulation to predict election results and if possible figure out discriminant variables. Which apparently contributed to John F. Kennedy’ s victory over Richard Nixon in 1960. Rather than admiring the analytic abilities of such precursors (!), the author is blaming them for election interference. A criticism that could apply to any kind of polling, properly or improperly conducted. The article also describes how Simulmatics went into advertising, econometrics and counter-insurgency, vainly trying to predict the occurence and location of riots (at home) and revolutions (abroad). And argues in a all-encompassing critique against any form of data-analytics applied to human behaviour. And praises the wisdom of 1968 protesters over current Silicon Valley researchers (whose bosses may have been among these 1968 protesters!)… (Stressing again that my comments come from reading and reacting to the above Nature article, not the book itself!)

da 5 bloods [film review]

I was most excited to see the new Spike Lee’s film, Da 5 Bloods, available on Netflix. As I had liked very much his much earlier films like She’s gotta have it, Do the right thing or Clockers. (Although I feel the original book had more impact, I felt.) But I was rather disappointed by this one. (Although I related with the few pictures taken at the War Remnants Museum in Ho-Chi-Minh City, which I visited in 2013!) As I felt it was wasting most of the story for the allegory… The heist story was implausible from start to end (which is admittedly an usual feature of heist stories), with the five guys going into the Vietnamese jungle on their own, 50 years later!, which makes them 70 years old at the very least, with a small back-pack each but enough to carry a complete metal detector, and finding gold and bones (not a true spoiler I think!), not worrying about mines (until it is too late). Some of the actors are terrific, especially the (PTSD) out-of-control Delroy Lindo who essentially carries the film and keeps it alive. But other characters remain dreadfully under-exploited, counter-productively for the story. Which (literally) implodes with too many divergent threads. All unraveling into botched conclusions and ending up into a mess of the movie, the message eventually shooting the messenger…

On top of this I also think the film is presenting a very one-dimensional view of Vietnam, from a postcard idyllic vision with buffaloes in rice paddies, to thugs working for a French crook. With the overused tropes of the faithful prostitute and the cigarette smoking femme fatale. Except the later is a propaganda speaker on the Vietcong radio and unlikely to smoke American cigarettes… And the 1950’s (pre-Điện Biên Phủ) attitude of the said French crook (including the “bad guy” Luger gun!) does not fit either. Of course, these anachronisms and clichés could be understood as a second degré choice, i.e. as a pastiche of earlier American Vietnam war movies, from Apocalypse Now (explicitly referenced at the beginning of the movie, copter, river boat trip and Khmer temple included) to The Deer Hunter (especially the Vietnamese xenophobia), to Rambo (with cartoonesque shooting scenes). Collating epoch newsreels with blurry and dreamlike recalls of the actual experience of the 4 veterans looking their present age is a stylistic choice, obviously, but its repetition does not help in creating structure or credence in the movie. Especially when the current day battles in the movie are not any further realistic, although intended to be so…

Charles M. Stein [1920-2016]

I have just heard that Charles Stein, Professor at Stanford University, passed away last night. Although the following image is definitely over-used, I truly feel this is the departure of a giant of statistics.  He has been deeply influential on the fields of probability and mathematical statistics, primarily in decision theory and approximation techniques. On the first field, he led to considerable changes in the perception of optimality by exhibiting the Stein phenomenon, where the aggregation of several admissible estimators of unrelated quantities may (and will) become inadmissible for the joint estimation of those quantities! Although the result can be explained by mathematical and statistical reasoning, it was still dubbed a paradox due to its counter-intuitive nature. More foundationally, it led to expose the ill-posed nature of frequentist optimality criteria and certainly contributed to the Bayesian renewal of the 1980’s, before the MCMC revolution. (It definitely contributed to my own move, as I started working on the Stein phenomenon during my thesis, before realising the fundamentally Bayesian nature of the domination results.)

“…the Bayesian point of view is often accompanied by an insistence that people ought to agree to a certain doctrine even without really knowing what this doctrine is.” (Statistical Science, 1986)

The second major contribution of Charles Stein was the introduction of a new technique for normal approximation that is now called the Stein method. It relies on a differential operator and produces estimates of approximation error in Central Limit theorems, even in dependent settings. While I am much less familiar with this aspect of Charles Stein’s work, I believe the impact it has had on the field is much more profound and durable than the Stein effect in Normal mean estimation.

(During the Vietnam War, he was quite active in the anti-war movement and the above picture from 2003 shows that his opinions had not shifted over time!) A giant truly has gone.