

BOOK REVIEW 

Year : 2011  Volume
: 7
 Issue : 1  Page : 100101 

Fooled by Randomness: The Hidden Role of Chance in Life and in the Markets
CR Sridhar
Operations Research, Fellow  Royal Statistical Society, President: Nuvis Anlytics Pvt Ltd, India
Date of Web Publication  5May2011 
Correspondence Address: C R Sridhar Operations Research, Fellow  Royal Statistical Society, President: Nuvis Anlytics Pvt Ltd India
Source of Support: None, Conflict of Interest: None  Check 
How to cite this article: Sridhar C R. Fooled by Randomness: The Hidden Role of Chance in Life and in the Markets. J Can Res Ther 2011;7:1001 
Author: Nassim Nicholas Taleb
Publisher: A Division of Random House, Inc., New York
Publication date: 2007
Price: Rs. 375
Random events, events that we have not foreseen, affect much of our daily life. When the effect of randomness is positive, we attribute the positive results to a specific skill that we possess, earlier unknown to us, and a negative effect to a random event. This attribution seems to be almost universal; both an idiot and an intelligent person exhibit this innate human quality. In his book titled "Fooled by randomness: The hidden role of chance in life and in the markets", Taleb discusses the concept of randomness and the refusal to understand its impact. Much of the material is sourced from his experience in the US bourses. However, the basic concepts that he expounds are of use to professionals in other trades too.
A random event is measured by its probability of its occurrence. Most probability estimates are based on observations. This is where the problems arise. A large number of observations do not necessarily lead to a strong hypothesis. The classic example cited by Taleb is that of a black swan; observing a million white swans is not good enough to lead us to a state that all swans are white. All it needs to decimate this assertion is the observation of one black swan. Why is the probability theory not applied widely in many areas of our lives? "Probability Theory is a young arrival in mathematics; probability applied to practice is almost nonexistent as a discipline."
When information is provided to us, it is important to figure out what noise is and what the signal is. When we are unable to do it, we base our decisions on noise leading to disastrous consequences. But how does one identify noise? Not easy!
We base our forecasts, quite often based on history. However, history can be quite useless in forecasting future events with accuracy. Taleb discusses several examples from trading to drive home the point. Risk is a known concept. The entire idea of risk arises when we are unable to predict an event with any certainty. It is fruitless to estimate risk with exact numbers, because the nature of randomness and the uselessness of history do not permit us to get an accurate numerical assessment of risk. It is more prudent to assess the risk and categorize it rather than estimate it and be fooled in to a false sense of security. The prediction based on historical data is through induction. Induction as a technique is a good tool in mathematics, but a bad idea to handle time series data in social sciences.
This leads to the problem of how one should assess an uncertain situation. Taleb strongly recommends Monte Carlo techniques to look at the sample paths that are available to make an assessment of risk. These set of techniques are effective tools to assess a random phenomenon. "The Monte Carlo man's realism is without the shallowness, combined with mathematician's intuitions without the excessive abstraction. For indeed this branch of mathematics is of immense practical useit does not present the same dryness commonly associated with mathematics". Monte Carlo in statistical sciences is analogous to the NewtonRaphson method in numerical analysis.
With enormous literature available on statistical sciences and huge computing resources, why do experts continue to make blunders in financial markets? Taleb offers a number of reasons in his book. I will discuss some of the reasons here.
Identifying a random occurrence with one's own skill is a major problem. Traders who make money in the shortterm refuse to accept that they made a killing due to a random event of very low probability. Instead they come up with jargon and equations to give the entire episode a scientific color. In the process, they forget to learn from random nature of markets.
Many including reputed economists refuse to learn that mathematical models with complex statistical equations do not reflect reality. Many of these professionals have pet theories and stick to them for life since letting these theories go will affect their professional and economic status. In this process, they block and ignore all the other paths that exist. After all "No theory can be ever right".
Real life consists of events that are distributed asymmetrically. With such distributions usage of standard statistics by scientists to make their lives easy leads to assessments that make no sense. When a model does not reflect reality jargon is used as a coverup.
It is impossible to observe rare events through observations. For example, if a surgical procedure continues to give consistently the same results a surgeon can never learn about a rare complication. The nonappearance of this occurrence will be taken as the absence of this rare event. This flawed deduction seems to be common and pervasive in medical practice. "Finding of absence and an absence of findings get mixed together" is a tool used by many drug researchers to twist a truth.
One of the most interesting concepts that Taleb expounds is that of Survivorship Bias. "We are trained to take advantage of information lying in front of our eyes, ignoring information that we do not see". Seeking and looking at information that we do not see needs a scientific mind of a tenacious individual searching for facts. "Remember nobody accepts randomness in his own success, only his failure".
A huge problem in inferential process is "nonlinear" effects. Typically, an effect where a small input leads to a disproportionate response is extremely difficult to model. The Sand Pile Effect is a classic illustration: a sand pile collapses when a last little handful of sand is added to a sand pile. There are more complex nonlinear processes with several variables involved in other real life situations.
Last but not the least is our mind. Antonio Damasio in his seminal book "Descarte's Error" demonstrates that a human devoid of emotions cannot decide on simple actions, such as drinking coffee. Our mind can deal with just one state at a time. "If your mind operates by a series of different disconnected rules, these may not be necessarily consistent with each other, and of they may still do the job locally, they will not do so globally."
In conclusion Taleb says, "The fact that your mind cannot retain and use everything you know at once is the cause of such biases. One central aspect of a heuristic is that it is blind to reasoning. Our brain functions by modules. An interesting aspect of modularity is that we may use different modules for different instances of the same problem, depending on the frame work in which it is presented."
In the last part of the book, Taleb goes more personal and confesses about his own inability to deal with randomness. The "three afterthoughts in the shower" nicely sum up his approach and experience of randomness in his professional and personal life.
"Fooled by randomness: The hidden role of chance in life and in the markets" is not a book that I would read once and store it in my bookshelf. There is a lot to ponder and reflect on his thesis of rare events. This book read in conjunction with Black Swan opens up new sample paths to our thinking. These two books have already achieved a cult status and have a large following.
