Showing posts with label Randomness. Show all posts
Showing posts with label Randomness. Show all posts

Friday, January 16, 2009

It Is Not Luck!


If you have spent a whole life time cursing your luck for all the bad things that happen to you or around you; or if you have lost faith in the system; or if you have no idea why your projects get delayed - you have come to the right place.

I will not be able to set things right, not give you a solution, but I can sure explain why things go wrong even when you are doing everything right.

But before that let us consider an entirely different scenario. Take a cup of water and drop a blob of ink (or any other dye). The ink disperses and soon the whole water is blue (or red or whatever). Of course you know why that happens. Diffusion. And that happens because of random motion of the molecules. So much is clear. But why diffuse? Randomness could also cause the ink to re-concentrate (imagine seeing the diffusion in a film that is run backwards - can you picture that?). But commonsense tells us that the ink will diffuse it will not come back together again. But that is not entirely correct. The probability of the ink 'un-diffusing' in not zero - except that it is so small that it never ever happens.

So it is with affairs around you. The probability of everything going smooth all the time and as you wish it to happen is not zero. It just so happens that the probability is so tiny that, left on its own, things will go wrong. But there is an essential difference between you and ink.


You can act. You can act to modify the probability in your favour.

Getting late to office because of the random occurrence of vehicles in your path and causing delay? Start earlier. You will still get delayed, but you will reach on time.

Projects get delayed no matter how well you plan? Some random occurrence of an event that you did not take into account manifested? Learn from this. Watch out for such random events. They will still happen. But if you are on a look out, you will be able to take appropriate action.

So it is not about luck. It is all about you trying to consciously modifying probability in your favour.

So next time you wish some one luck, try saying, "Wish you favourable probability." If nothing else I guarantee you raised eye brows.

Oh come on! What is life without such a bit of Idiosyncrasy?

Note: The picture used belongs to Steve Woods (See gallery)

Stumble Upon Toolbar

Monday, December 29, 2008

When Cherished Ideas Crash

Ever had a feeling that your carefully constructed world view is crashing all around you as you read a more potent argument against it. You then scramble to qualify your world view so that it remains intact because you have seen it work before - but you are not so sure. Of course you can always have a fall back option (see Fickle) but not before giving a fight.

My basis for determining the validity of any information has always been to evaluate the information against a context. I keep asking, "what is the context?"  So, if some one says, "such and such HR initiative has worked in, say, Infosys." I would always ask, "Let's first examine the context."

And if the context is similar to mine, I would adopt it. To me context represents the sum of all independent variables that matters.

Similarly, story telling, to me at least, has always been the best way of passing information. But apparently narrating is too simplistic a tool and has the impact of forcing certain post-hoc causality on an event where there might be none or many that escapes detection.

Reading The Black Swan and before that Fooled By Randomness has been an humbling experience.

Context matters, but only up to certain extent. Actually only up to a small extent. Ditto with evidence. Because for every evidence and/or context there are many more that are not obvious or hidden.

No I have not given up on my context based filtering of information, but what does one do to counter this:

I propose that if you want a simple step to a higher form of life, ..., then you may have to denarrate, that is shut down the television set, minimize time spent reading newspapers, ignore the blogs. Train your reasoning abilities to control your decisions ... This insulation from the toxicity of the world will have an additional benefit: it will improve your well-being.


Mr. Taleb, would you reconsider leaving "ignore the blogs" bit out from the above quote? We actually help you detect black swans! How? By presenting counter points. If that does not help detect black swans, nothing will.

Stumble Upon Toolbar

Thursday, October 23, 2008

Always Come Early For An Interview

the long job interview queue
Or how to improve your probability ...

I cannot get over this book. Fooled By Randomness so vividly describes how we human do not understand how randomness fools us.

One fact that emerges out of this book is that traders take probability of an event occurring as risk while actually it is the value associated with the risk that needs to be considered.

I therefore decided to test out this hypothesis.

I asked a family member if she would place a bet on a game if there was 99% probability of winning Rs.100 but only 1% probability of losing Rs.100,000. The answer came almost immediately, "yes". I then asked her the second question. Would she undergo an eye operation which has 99% probability of perfect eyesight post-operation but only 1% chances of losing the eye. "Are you crazy?" was the answer.

My conclusion: people understand risk when the value is more personal. The first question remains in the realm of the hypothetical. Though the value is negative, there is actually no such game and she would not lose any money since we are not playing one. But an eye? We understand the value of losing an eye. That is more immediate. I am not a trader but I can now guess why brokers do not worry about value while taking risk. It is not their money - most of the time. The thought of a loss does not hit them hard enough as the thought of losing an eye would. Therein lies the difference.

But wait there is more.

The above incidence happened a few days ago. Yesterday a colleague asked me if I have seen the movie "21". She then asked me why switching the choice of the closed door in the game show would result in 67% probability of winning, when it is obvious that the probability is 50%. (In the movie the professor asks the protagonist the famous Monty Hall Problem, in which the game host offers the player a chance to switch the door after he has revealed that one of the doors not chosen does not contain a valuable item.)

If you follow the this link you will get an answer. But what is important to understand is that not only do human beings react to probability in a very simplistic way but we also refuse to believe that an action by someone could change the probability, especially when that person is in possession of some knowledge that you do not have.

We tend to treat all events as independent. Just like the previous roll of dice has no impact on the next roll of dice. But that is not correct with life except for some very simplistic cases. If you are appearing for an interview in a job, assuming it is a fair selection process, do you think that every candidate has equal chance of selection? Wrong. Your chance depends on who has gone before you. If the candidate before you impresses the selection committee, unless you are superlative, your chances reduce.

Note: The picture used here belongs to Sigurd Decroos. To see more of his pictures go to his gallery here.

Stumble Upon Toolbar

Monday, September 22, 2008

Why do I buy books

Old Books
I always read books twice. If possible within 6 months. I have experienced that re-reading a book within that duration ensures long term memory retention. That is why I need to buy books. So every book that I discuss on this blog has been purchased by me. Some on Amazon but most during trips. Nothing beats reading something interesting while waiting at airports.

Right now I am re-reading Fooled By Randomness by Nassim Nicholas Taleb. It has been dubbed "One of the smartest books of all time' by Fortune. I would agree, although I cannot claim any mastery over economics and finance.

The following extracts from the preface of the 2nd edition will tell you what to expect:

"Past events will always look less random that they were (it is called hindsight bias)."

"... we don't have much of a clue in the social "sciences" like economics in spite of the fanfares of experts."

"... the kind of luck in finance is of the kind nobody understands but most operators think they understand ..."

This is just the book I need to read (re-read?) to put the Wall St. happenings in perspective.

Note: The photograph used belongs to Zsuzsanna Kilián. Please go here to see more such photographs.

Stumble Upon Toolbar

My Library