Jump to content
Objectivism Online Forum

The Black Swan

Rate this topic


earwax

Recommended Posts

I am just making my way through this book now. It is entertaining enough so far not to lose my attention. I was wondering if anyone else here had read the book and had any thoughts on the subject. I am not finished with it so I am not sure where he is going to end up. I am at the part where he is discussing what he calls 'the error of confirmation.' I am not quite sure where he is going to end up. He seems give to much weight to the chaos in the world then I would prefer but he does seem to have some valid points on the limits of inductive logic.

Link to comment
Share on other sites

I am just making my way through this book now. It is entertaining enough so far not to lose my attention. I was wondering if anyone else here had read the book and had any thoughts on the subject. I am not finished with it so I am not sure where he is going to end up. I am at the part where he is discussing what he calls 'the error of confirmation.' I am not quite sure where he is going to end up. He seems give to much weight to the chaos in the world then I would prefer but he does seem to have some valid points on the limits of inductive logic.

The book is based on Poppers black swan example . He is building off of his skepticism of induction. I recommend reading POPPER AND AFTER by Stove and Bo Dragsdals KARL POPPERS ASSAULT ON SCIENCE. to see the problems with Poppers theory.

Link to comment
Share on other sites

How would you characterize Stove and Gragsdals' philosophy? I generally think Popper was on the right track with his move to focus on "what one ought to believe". Mix in some Wittgenstein on "what is truth" and I think you've got a recipe for success.

Link to comment
Share on other sites

Its not scepticism towards induction, its a study of how rational decision-making changes when you introduce a high penalty for being wrong. In day-to-day life relying on inductive judgement is fine because even though theres always a possibility that youre wrong, this possibility is normally quite small and the costs for being incorret arent too high. If I'm a scientist studying swans and every swan I've ever seen is white then its logical for me to hypothesize that all swans are white - yeah, its possible that one day a black swan will turn up and disprove my theory but who cares? Scientific theories exist to be revised and theres nothing bad about saying "ok I was wrong" and updating the theory.

However in finance things are slightly difference because even if the possibility of being wrong is extremely small, the cost that you incur if your model is incorrect can be astronomical. If I'm a hedge fund using a statistical model which is 'wrong' yet works 99.9999% of the time then I cant just ignore the 0.0001% chance of the model failing because if it does I might lose everything I have. So a proper risk management strategy needs to be robust against the possibility that the model I'm using is fundamentally wrong, purely because the costs that I can incur if this is the case are so high. Decision making isnt just about maximising the probability of being right, its about balancing risks, and sometimes the decision which is optimal if youre only concerned about being right becomes suboptimal once you start factoring in the relative costs that you incur if a mistake is made. Theres no point being right 99% of the time if the benefits from being right are outweighed by the loss you suffer during the 1% of the time your predictions are incorrect.

His other key point is that its very difficult to get an accurate estimate of the costs/probability of being wrong, since calculating these will generally presuppose some kind of model and this itself may be incorrect. So standard risk management strategies will often underestimate the likelihood/costs of mistakes (Goldman Sachs classic statement that their losses during the subprime crisis "were a 25 sigma event" and so on) because its difficult for a model to predict the consequences of its own failure without making assumptions that are hard to justify. Any attempt to model abnormal conditions is subject to scepticism.

As for the book, I think most of the points he makes are correct (some to the point of being trivial), but his smug tone really annoys me. Theres not much in the book which isnt obvious to a practicising statistician but he continually has the attitude that he is the only person that understands the things under discussion while the scientific community is labouring under misconceptions. The correct conclusion to draw isnt that statisticians dont understand the logic of decision making, it's that the pressures of working in a some environments (eg financial) often causes people to ignore their better judgement for the pupose of making short-term gains. So I'd say that the book is more a study of greed than anything else.

Edited by eriatarka
Link to comment
Share on other sites

His other key point is that its very difficult to get an accurate estimate of the costs/probability of being wrong, since calculating these will generally presuppose some kind of model and this itself may be incorrect.

This reminds me of another book I've been thinking of getting, Stumbling on Happiness. The basic point of the book is that we are not very good at estimating how something will make us feel, and consequently, we put a lot of stock in a future decision, which won't have a great effect on us. For example, thinking that brand new pair of trousers or new mp3 player. Conversely, we tend to make the best of the situations we are stuck in and find ourselves, as the book-title suggests, stumbling on happiness.

The point of the book is supposed to be not that life is unpredictable and random, but that we need to learn how to maximise our ability to see the value of what we have and to critically evaluate what we don't have.

A short review of the book that interested me.

Edited by Tenure
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...