Enter your mobile number or email address below and we'll send you a link to download the free Kindle App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter your mobile phone number.
Stalking the Black Swan: Research and Decision Making in a World of Extreme Volatility (Columbia Business School Publishing) Hardcover – March 31, 2010
|New from||Used from|
Best Books of the Year So Far in fiction, nonfiction, mysteries, children's books, and much more.
Customers Who Bought This Item Also Bought
As a respected equity analyst, Kenneth A. Posner has seen firsthand what it means to be blindsided by black swans. In this deeply thoughtful and heartfelt book, he traces out the sources of what we think of as 'surprise events' and offers a pragmatic approach for confronting and hopefully mitigating their adverse effects. Along the way, there are many beautifully elaborated stories depicting how swans often won the day, as well as descriptions of the all-too-rare positive outcomes. This is a valuable book for anyone who treads a path through an uncertain world (hint: that should be everyone!).(Martin Leibowitz, managing director, Morgan Stanley, and coauthor of The Endowment Model of Investing)
How do highly intelligent bankers and investors get surprised and brought down by financial crises created by their own interactions? Why can't formulaic rules and regulations prevent such crises? How can sophisticated models be so far off? How can one cope with the 'the unpredictability of collective action'? Posner provides insightful, honest, and very instructive real-world adventures in how 'to learn to live with extreme volatility,' as financial markets marched, trumpets blaring, into the quicksands of recursiveness.(Alex J. Pollock, American Enterprise Institute, and former president and CEO of the Federal Home Loan Bank of Chicago)
Posner's Stalking the Black Swan is an insightful integration of the emerging field of behavioral economics with real-world insights about financial markets. Posner combines true intellect, a fascinating professional background, and a critical mind to help us understand where laboratory insights hold up and where they might not. This book will be a valuable read for anyone interested in the role of behavioral economics in predicting market behavior.(Max H. Bazerman, Straus Professor, Harvard Business School, and author of Negotiation Genius (with Deepak Malhotra) and Judgment in Managerial Decision Making (with Don A. Moore))
Posner has drawn on his extensive experience as a first-rate analyst to examine the clues leading up to the recent meltdown of a number of once-respected financial institutions. But this is not a financial retrospective. He shows us how to combine numbers crunching with intuition, and he provides us with a set of disciplines to use in anticipating future Black Swan eventsboth positive and negativeand to profit accordingly.(Byron R. Wien, Blackstone Advisory Partners LP)
If you are looking for a well-written book which seeks to merge quantitative analysis, fundamental research, and human psychology, Stalking the Black Swanprovides a great resource to help you fine tune your investing framework for the volatile environment of our times.(Wall Street Cheat Sheet)
a wonderful read on how to think about, analyze, and react to extreme events.(Motley Fool)
Stalking the Black Swan provides a great resource to help you fine tune your investing framework for the volatile environment of our times.(Damien Hoffman Wall St. Cheat Sheet)
Stalking the Black Swan doesn't attempt to explain why or how the crisis happened, but instead offers practical, crisis-tested advice on methods for uncovering a similar looming disaster in your own portfolio.(Seeking Alpha)
A unique combination of psychology, statistics, and economics that will engage the reader unwilling to settle for quick answers.(Aswath Damodaran, New York University)
Top Customer Reviews
So what is so unique about Ken's investing philosophy? The concepts of value investing, information assimilation, behavioral finance, and probabilistic thinking are certainly not new - but the way he effectively stitches them together into an easy to understand and apply framework is. Most importantly, Ken's framework is far from static, and in my view, the biggest accomplishment of this book is that the reader comes away with a basic understanding of how to adapt Ken's framework based on the different analytic needs of each stock, rather than applying it mindlessly in checklist fashion. So while I view most other investing books as individual "tools" that investors can add to their investing "toolbox", Ken's book is the entire toolbox along with embedded knowledge on how to select the appropriate set of tools for each different job.
Ken also tackles core issues in investing that are frequently neglected by most popular investing books such as confidence/conviction, how to handle information overload, and how to ask management questions to maximize the information you get from their answers (or non-answers) - the last topic is well worth the price of the book alone.
This book is a must-read for any investor that wants a sustainable edge in this business. While it takes a lot of hard work and discipline to follow Ken's philosophy, I am convinced that adhering to the core concepts of this book will maximize the reader's chance of having a long and successful investing career.
He spends an entire chapter (8) on Monte Carlo simulation. But, he does not understand the basics of it. He states that Monte Carlo simulation generates probability trees with thousands or even millions of branches (pg. 178). No, that's not the case. A Monte Carlo simulation structured that way would never get off the ground. Even really powerful computers would most probably freeze because of the excess computational burden associated with billions of calculations. Instead, Monte Carlo simulation does not add one single branch to the original mapping of your model. But, it can change the probability of each branch (by specifying a distribution) and the input value of each branch (by turning variables into random ones). And, it can run millions of iterations. That's completely different and a lot less burdensome than creating a million of branches, as he states.
He mentions Principal Component Analysis (PCA) (pg. 191) and he states that PCA can map out the correlations in systems with large numbers of variables. No, it does not. Instead, it simply allows you to overcome multicollinnearity issues that often surface when you deal with a lot of variables. That's a different thing. Instead, you achieve what he is talking about with Pathway Analysis or a correlation matrix.
Besides misunderstanding quantitative methods, he makes numerous questionable statements. On pg. 21, he explains the bankruptcy of Bear Stearns, the failure of AIG, and Wachovia due to abrupt fall in stock prices. But, the latter is not the causal factor (stock price drop), it is the effect. Bear Stearns failed because it could not support two huge hedge funds it sponsored that collapsed when its MBS and CDOs portfolios lost much of their value at the onset of the subprime crisis. AIG failed because it issued a huge amount of CDSs without reselling the related risk to anyone else. Wachovia failed because of material credit and asset quality problems in all its loan portfolios. Those were the causes of their respective failure. And, traders responded to those developments and their respective stock values crashed.
On page 77, he states that Bank of America's acquisition of Countrywide was a great deal. No, it was a disaster. BofA had to weather $billions in asset quality write downs, investor lawsuits, and regulatory penalties related to its Countrywide acquisition. Some of those risks had well materialized by the time Posner published this book in 2010.
On page 125, he states a test of diagnostic power is the ratio between true positives and false positives. Instead, it is the ratio between true positives (numerator) and the sum of true and false positive (denominator).
On page 162, he indicates that when you add more variables to a model, some of the random errors should cancel out as a benefit of diversification. But, the proper framework to capturing the benefit of adding variables to a model would be focusing on its Adjusted R Square or its Standard Error. Using those metrics, you readily observe that adding variables beyond the first four or so runs into diminishing returns. That is especially if you further test your model by using Hold out samples to filter for overfit models.
On page 167, he interprets that a stock that moved by 300% over 12 months with an implied volatility of 30% resulted in a move of 5-standard deviations. Instead, the answer is 10-standard deviations (300%/30%).
On page 199-202, he explains how selling Puts with a strike price of $55 would have made much money when the stock price recovered and passed that level on the way up. But, that's not possible because those Puts would have been exercised a long time ago when the stock fell to $35.
Within chapter 9, he deems that the quant strategies failed during 2007 and so did value investing strategies. But, he never provides any comparisons. Without relative comparisons between different strategies, his statement provides no information or assessment of those strategies.
His main tool, a probability decision tree, has very limited predictive power. This is because the vast majority of his probabilities within the branches of such models are purely subjective. His probability trees are different from Bayesian trees. Within Bayesian trees you updated probabilities with actual new information. You start with subjective probabilities, but you update them with true frequencies. Instead, Posner never has true historical probabilities to update his decision tree model with. He just arbitrarily changes them. But, such changes in probabilities are entirely subjective. It appears the vast majority of the time, his probability decision trees did not help him in predicting the correct direction of prospective stock prices.
After the fact, Posner explains his failing of his predictions by stating I should have weighted more this factor or this one. Or I should have known about this "lurking" causative variable. But, that is a futile exercise in 20/20 hindsight. If beforehand, we could figure out the exact causal variables to focus on, weigh them accurately, figure their impact equally accurately, and forecast the future path of the economy (or at least correctly estimate the relevant probabilities of different economic path)... then Posner's framework would be effective in making predictions. But, we can't do any of that. To further explore why we can't I recommend the excellent book by Duncan Watts Everything Is Obvious: How Common Sense Fails Us.
On page 10, Posner indicates how much improvement weather forecasting has made over the past few decades. Meanwhile, economic forecasting accuracy has shown no improvements over the past 30 years. Nate Silver in The Signal and the Noise: Why So Many Predictions Fail -- but Some Don't will make the exact same remark two years later. However, he will add also the fields of financial and security analyses to the fields that have shown little improvement in forecasting accuracy. It makes good sense, since security analyses forecasts are very dependent on economic forecasts.