Friday, February 18, 2011

Predicting the future

Wouldn't it be great if we know what is going to happen tomorrow, next week, next month?

It is such a tantalizing thought that we try to predict the future all the time.

Subconsciously, we predict when our foot will land when we take a step and we get surprised if the landing does not fit our expectations.  Ever walked up and down stairs that did not seem right?  You were predicting before finishing your step.

On a more conscious level, we predict the future when we plan, form an expectation of what is to come, and actively arrange things to move future events towards what we would like to see.

In fact,  one definition of intelligence is the ability to quickly learn a new situation and accurately predict what the outcome maybe.  Therefore, an intelligent person is able to get to the core of a game not only in terms of the rules but also the strategies to win faster than those less intelligent and predict more accurately what the outcome of moves in the game will result in.

The problem is that even though we do it all the time,  we are always struck by the times when our prediction is wrong.  If only we can be more accurate, we can rely on it.

What are the causes of these wrong predictions?

Let's look at an example I read from some notes on logic.

You enter a room and find a transparent glass jar with different colour candy in it.  There are red ones, orange ones, as well as yellow, green, and black.

You try a red one and find it to be cherry flavoured.

Do you think that if you try another red candy that it will be cherry flavoured?

Most of us would say yes but the logicians would say no.

Their view is that trying to figure out the flavour of the next candy is foretelling what is going to happen in the future which is impossible.

Just because the sun rose and set today does not guarantee that it will do it again tomorrow.

People who attributed good luck to charms and rabbits foot can make wrong predictions if the prediction is based on whether they have their lucky charms with them or not.

There is actually a term call the "turkey illusion".  To a turkey growing up in a turkey farm,  the turkey farmer is his best friend because everyday the turkey farmer brings food and clean out the pan.  Can the turkey expect this to go on forever base on past experiences?

A few weeks before thanksgiving,  the turkey farmer came to wring the turkey's neck.

However,  the logicians will concede that you will be able to predict candy flavour if you know the rule that the flavour of the candy is uniquely tied to the colour of the candy.

But of course,  that is what we tend to "assume" which can be right a lot of the time but not all the time.

Logicians would like to be precise and examine every premise before making a conclusion.

Real life demands that we make assumptions or else nothing gets done.

We therefore make predictions based on assumptions that can be mostly true but often overlook these assumptions.  If we add oversights and missed information to this mix,  we can see why it is quite difficult to predict the future, especially on situations that we don't have much experienced to draw from.

We have all seen the fine print "past performance is no guarantee of future returns" in investment advertising while loudly proclaiming how well they have performed in the past.

"Turkey illusion" warning while telling us to be practical and make an investment instead of leaving money in boring bank accounts?

The candy example also reinforce my dissatisfaction with philosophical explanations expressed in my earlier post on life.  If we can solve the prediction of candy flavours with a rule tying flavours to candy colours,  we have only shifted the uncertainty to the rule of candy colours.

We now need another rule to say that this rule for candy colours will not change tomorrow or in the future,  which we can never be sure of.

So again,  we seem to be making progress in understanding but truth just backed off a bit further.  Sometimes, it seems to back off even further than the progress we just made!

No comments:

Post a Comment