Tuesday, July 01, 2008

The Black Swan – Nassim Nicolas Taleb

I have fond memories of this author – one of my favourites among the books I’ve read. I was hooked by his style of skeptical empiricism when I was introduced to his first book – ‘Fooled by Randomness’ – by serendipity. The author will be glad that it is through a series of black swan event that lead me to his first book, and consequently, to his second book.



Long time ago, my gf went to US for a conference on a paper she submitted. Man, in these conferences, there are people literally giving out books for free (to be fair, they did try to sell during the first few days of the conference, but towards the last day, book hell went loose). My gf, a typical free-must-grab Singaporean, grabbed a few of the books, regardless of race, language or religion (ok, I exaggerate). Among the horde of books, one of them is Nassim Nicolas Taleb’s ’Fooled by randomness’.

Talk about Black swan events (these are incalculable, low probabilistic and highly consequential events), my introduction to the author’s first book must fall squarely into such category. The author’s style is very refreshing, a logical salad mixed with lots of stories woven with facts, creating a potent mix of philosophical brouhaha.

Black swans are named as such because of this story. Imagine all throughout your life, you only saw swans which are white. Based on historical past, you can ‘extrapolate’ your data by induction that all swans are white. This is all jolly well and good until one day you saw your first black swan. This event totally tears away your hypothesis that all swans are white and is something that the past data can never predict. The probability of meeting a black swan is not calculable, though based on the past data it has a very low probability (which explains why you didn’t see one earlier) and has serious consequences. Here, ‘absence of evidence’ is misconstrued as ‘evidence of absence’.

Are we similarly fooled by such logic errors? I can think of a few:

1. From analyzing a company’s past earnings for say 10 years, I come to the conclusion that the earnings are stable and growing steadily. I treat the absence of evidence of ‘poor earning years’ as the evidence of absence of ‘poor earning years’, leading to my skewed over-bullishness of the company.

2. A man, having lived till a ripe age of 100 years old, declares that from his 100 years of non-dying, he is an immortal and thus will carry on living for a few 100 years more. He made a mistake of thinking that ‘no evidence of death’ is the same as the ‘evidence of no death’.

These are exaggerated examples to illustrate the points, but it blows my mind to think in this way.

‘The Black Swan’ carries with this style of writing, perhaps more brilliantly so. The new book talks about how we systemically and biologically ignore black swan events. In fact, he argued that the world is governed mainly by black swan events, not by regular and inconsequential events that are predictable. He goes as far as to say that what we do not know is far more important than what we know. Among the most important things I carried away from reading the book, is this notion of ‘silent evidence’ – how we are blinded by things that are not found in the sample size, hence we discounted them to the extent of skewing our perception of things.

Below is an example to illustrate this point:

Consider the world’s richest people – all of them exhibit traits of risk-taking, go-getter mindset, determined etc etc. Thus,

Rich people have a certain characteristics.
I have these characteristics.
Therefore, I’ll be a rich person.

This is logically flawed. Having a set of characteristics exhibited by rich people does not necessarily mean I’ll be a rich person. There are many examples of people with these characteristics but are not necessarily rich. Hence, these people who had these set of characteristics could be there based on pure luck.

I think these ideas ties in very much with the idea of causality. Correlation does not necessarily imply causality, though causality implies correlation. For example, almost 99.9% of cancer patients drink water (high correlation) but it does not mean that drinking water will cause cancer (no causality). We are often tricked into believing such logic errors, which tend to exacerbate the consequences of black swan events.

------------

The review is split into two parts because the books is exceedingly thick and I had to write down my thoughts halfway before it gets diluted and lost in transition.

This had to be the most the most relevant book I’ve read in my life. It’s like I’ve been born blind, then after a corrective surgery, I see the world as it is now but had been elusive to me before. I can never look at the world again because the book had permanently changed me. Don’t go out and grab the book because I said so…my experiences lead me to different perspectives when I read the book, which will be vastly different from another reader.

The second half of the books talks about more technical stuff. I confess I do not really understand the whole of it in the first reading. Perhaps, like Intelligent investor, the true gist of it will only be realized after subsequent readings. The author highlights the fallacy of treating everyday occurrences to fit the bell-curve (the Standard normal or Gaussian distribution). This fallacy is so rampant that it permeates most of the ‘scientific’ methods regarding a range of socio-economic disciplines like economics, sociology and finance. I’ve never believed in the Modern Portfolio Theory (MPT) and efficient market hypothesis (EMT), so this didn’t struck me hard enough to feel defensive about his ideas.

I guess the author wants us to follow a bottom-up approach rather than a top-down approach in everything. A bottom-up approach means using real, empirical data to look at the world. A top-down approach means learning ‘scientific’ theories and principles, then fitting data that follows them and ignoring or downplaying data that do not.

A simple example would be a stereotypical Western trained doctor who abhor using traditional chinese methods like acupuncture or herbs, judging it unscientific (though it’s very empirically based) because of top-down approach.

Am I also blinded by my own theoretical framework?

The author suggests a few ways to deal with the uncertain world:

1. Make a distinction between positive black swans and negative black swans. Black swans events can benefit or cripple. Once identified, expose yourself maximally from positive black swans, and limit exposure to negative ones

2. Invest in preparedness, not in prediction

3. Seize any opportunity, even those that looks like opportunity. This is the same as exposing one to positive black swans. Work hard, not in grunt work, but in maximizing one’s exposure to opportunities.

4. When caught between unknown probabilities of choices presented to you, focus on the (known) consequences of each choice and not on the probabilities. It’s also known as Pascal’s Wager.

I started this book being blind, and I finished this book knowing that I will never view the world the same again.

11 comments :

Anonymous said...

Wow you seem like a really profound person. Are you a professor? May I ask how old are you?

Btw I love your blog! There's so much to learn from you. Pls keep blogging :)

Anonymous said...

thanks for the review, i have been wanting to borrow from library (without much success).

it seems so hard to grasp! And are his theorems gives useful application in daily life? sorry, I don't see much usefulness. maybe I'm blinded by theoretical framework as well?

anyway, just sharing my mediocre thoughts & trying to relate what already existed for so long.

1. Make a distinction between positive black swans and negative black swans
...........
if you short on bear, long on bull - you will ultimately benefit. so essentially just need to be at the right side at right time?

2. Invest in preparedness, not in prediction
.............
I think he did mention about putting more in predictable funds (bonds etc) while putting some in speculative ventures (options). Well, that's similar to our basic Asset Allocation doesn't it? i.e to spread our risk which also prepared us for the unpredictable future

3. Seize any opportunity, even those that looks like opportunity.
...........
hmm ... i thought this is obvious? of coz we will need to apply "margin of safety" when assessing opportunities.

4. When caught between unknown probabilities of choices presented to you, focus on the (known) consequences of each choice and not on the probabilities
........
well, this akin to saying when future becomes uncertain, move all money into saving account :)

oh boy, i read that he did provoke a lot of people! LOLz

la papillion said...

Hi little,

Hoho! Dun be fooled by me :) My age is the product of 5 and the square root of the product of 18 and the smallest prime number :) I had the misfortune of nearly becoming a phD, hoho!

Thks for your kind comments :)

la papillion said...

Hi durio,

He doesn't have theorems in the books. He's just taking a book to discuss on the logic errors and about people being fooled by randomness.

Was it useful? It is, tremendously. It opened my eyes to another world, a world where it's okay to say 'I don't know' and a world not bounded by (illusory) certainties. To follow his ideas is to be skeptical of the certainties of theories and principals, to let empirical data form your own framework of how the world works.

It is philosophically life changing for me.

Let me comment more about how I find his book useful for myself in my future blog articles. It'll be more respectful to you (after you had spent effort in writing so much) than giving some touch-and-go comments here.

PanzerGrenadier said...

Hi La Papillon

You have made me hunger for this book. :-)

A exceptional book review as you have summarised it very simply for us on what the book is about and added your own views on how it helped you.

Would you like to do a link exchange with my blog? We are both featured on thefinance.sg but I'd like to do link building with like-minded fellow personal finance and investment bloggers in the Lion City. ;-)

Panzer

P.S. I like your humour too...

la papillion said...

Hi PG,

I've been an avid reader of your blog too :) Thks for your kind comments! I'll add you in my links :)

Anonymous said...

Hi LP,

Read "Fooled by Randomness", ironically picked it up randomly from the library.

Was just thinking about the increasing demand for "risk managers" in fund houses after the credit crisis.

Based on statistical extrapolation, the safest point for these guys was the moment before the credit crisis.

I think it would be insane to accept such an impossible job...

Financial Journalist said...

I recommend this book: Traders, guns and money.

This is another great book, read it, you will not regret.

la papillion said...

Hi wisdomw,

Yea, I totally agree. How to be wary of the risk that one knows nothing about? How to be careful of the unknown unknowns? It's a fool's errand.

Anonymous said...

LP,
you're too kind!
looking forward to more articles on the black swan. thanks!

PanzerGrenadier said...

Hi La Papillion

I've written my review of Nassim Nicholas Taleb's "The Black Swan"

http://fivecentstencents.com/2008/10/12/how-to-avoid-getting-killed-by-a-black-swan/

Thanks for reviewing the book earlier so that it made me want to read it. One of the top reads for 2008 for me IMHO. :-)

Be well and prosper.