These days, predictions fall in one of the 3 categories:
- Predict all possible outcomes and publicize the one that goes right.
- Predict the outcome so far in the future that there is no way to hold you accountable.
- If your predictions don't seem to go your way, just change them.
I find most of the current financial predictions related to GDP growth, inflation, RoI on investments, sporting outcomes including cricket matches and even weather forecasts follow the above trend. Most of the so-called pundits evaluate all kinds of data and analyze situations to arrive at seemingly concrete predictions about the outcome. How many times have we seen elaborate reports from consulting firms, investment banks, research companies, etc predict 'accurately' the growth of companies, even countries, with all the complex calculations and numerical wizardry. These days we even see experts on television channels engage in predicting events ranging from how much will it rain during the week to who is likely to win the latest T20 match between India and New Zealand.
Now, I am all for quantitative analysis and modelling to arrive at a logical equation that predicts the output. Basic forms include Y = f(x,y,z..) equations where x,y,z... are the variables on which the outcome (Y) is dependent upon. The equation is fairly straightforward as long the variables x,y,z... can be identified and the relationship among them established. However, things begin to get murky once the list of variables increases and the relationship among them becomes complex. Situation gets out of control when it becomes difficult to even identify the variables let alone establishing the way they interact with one another. This is precisely the case with most of the complex predictions doing the rounds today. Things like GDP growth, inflation or even sporting outcomes are dependent on so many variables that it is impossible to take into account the role each plays and an outlier can change the outcome dramatically. For example, not so long ago when people took the 8-9% growth rate of China and India for granted, little did they realize the effect of a recession in US and the debt crises in Europe or even the complex political issues in India and the labor issues in China. Now that things are not as per expectations, there is doom predicted all over and it seems that 5% is the new 'normal' in India. The reality is often somewhere in between.
The central point being it is extremely difficult, if not impossible to predict or even establish general trends to things of such complex nature. At best, it can be an informative guess which comes with all the required disclaimers. For people who might think otherwise, one can always pursue one of the course of actions mentioned earlier.
- Predict all possible outcomes and publicize the one that goes right.
- Predict the outcome so far in the future that there is no way to hold you accountable.
- If your predictions don't seem to go your way, just change them.
I find most of the current financial predictions related to GDP growth, inflation, RoI on investments, sporting outcomes including cricket matches and even weather forecasts follow the above trend. Most of the so-called pundits evaluate all kinds of data and analyze situations to arrive at seemingly concrete predictions about the outcome. How many times have we seen elaborate reports from consulting firms, investment banks, research companies, etc predict 'accurately' the growth of companies, even countries, with all the complex calculations and numerical wizardry. These days we even see experts on television channels engage in predicting events ranging from how much will it rain during the week to who is likely to win the latest T20 match between India and New Zealand.
Now, I am all for quantitative analysis and modelling to arrive at a logical equation that predicts the output. Basic forms include Y = f(x,y,z..) equations where x,y,z... are the variables on which the outcome (Y) is dependent upon. The equation is fairly straightforward as long the variables x,y,z... can be identified and the relationship among them established. However, things begin to get murky once the list of variables increases and the relationship among them becomes complex. Situation gets out of control when it becomes difficult to even identify the variables let alone establishing the way they interact with one another. This is precisely the case with most of the complex predictions doing the rounds today. Things like GDP growth, inflation or even sporting outcomes are dependent on so many variables that it is impossible to take into account the role each plays and an outlier can change the outcome dramatically. For example, not so long ago when people took the 8-9% growth rate of China and India for granted, little did they realize the effect of a recession in US and the debt crises in Europe or even the complex political issues in India and the labor issues in China. Now that things are not as per expectations, there is doom predicted all over and it seems that 5% is the new 'normal' in India. The reality is often somewhere in between.
The central point being it is extremely difficult, if not impossible to predict or even establish general trends to things of such complex nature. At best, it can be an informative guess which comes with all the required disclaimers. For people who might think otherwise, one can always pursue one of the course of actions mentioned earlier.
No comments:
Post a Comment