Taoyuan Nights

Easy predictions & publicity-seeking behaviour

Another company is desperate to make a name for themselves again, after their absolute failure to predict any of the recent events that have plagued Taiwan in the last year. Here, they’re making a ‘prediction’ that has an extremely high chance of coming true simply by chance. And even then, they have to add the word ‘may’ at the start – making this statement 100% likely to be true, no matter what happens.

In contrast, real predictions can be falsified; and real predictions are for events that are unlikely to occur simply by chance.

Correlation is not Causation; abuse of the word ‘as’

I’ve talked about this kind of media junk in my last post. This time, I want to draw your attention to something else (something that Michael Turton is famed for highlighting on his own blog). Look at this sentence:

“The Taiex surged the most in eight years on April 30 as the island allowed Chinese investment for the first time since the end of a civil war in 1949.”

You see how they used the word ‘as’ and placed these two independent things in the same sentence, so as to imply a connection between them? As though A caused B? But how do we know it did? This ‘Chinese investment’ issue has been an ongoing issue for the last year or two now with continuous small movements in that direction.

I mean, “I stood up as a car drove past”. That hardly means the two are connected; but the financial media of the world are wont to abuse the connective ‘as’ in this manner. It’s worthwhile to be aware of this: two things happening around the same day, does not mean they are in any way connected, despite what the financial media would like you to think. Look out for the word ‘as‘ being abused, next time you watch a financial program on TV or read an article about finance or the economy.

In this case, on April 29th (US time), the DJIA had jumped 200 pts – a fairly steep rise – which would have had an impact on the Asian markets as they opened 8 hours later. Why isn’t this mentioned? It would seem to be very likely to be a contributing factor, since it is fairly well established that drops and rises in one major equity market tend to cause similar behaviour in other equity markets around the world (it’s something like when a supermarket drops the price of beer; you may find that other supermarkets nearby do exactly the same thing to compete).

Cargo Cults

The reason why this particular article was written goes a little bit deeper than mere page-filling and the abuse of coincidental events: this is really about the Cargo Cult effect. It is common to see this in the Taiwanese media (and media relating to Taiwan); a mystical belief that anything to do with the mainland and investment will magically lead to riches for Taiwan.

However, there’s no reason to believe this is the case. It’s my understanding that a reasonable part of Korea’s economy was hollowed out by China, and the Japanese have had to take careful steps to prevent the same thing happening to them too. The best place to read more about this is Michael’s blog; I’m just highlighting it here for anyone that hasn’t heard of this problem in the financial media coverage of Taiwan. China may be a positive thing for Taiwan, or it may be a net negative. But these days, you will very seldom find coverage of the negative angle in the Taiwanese media. Most people want to believe that a saviour is coming to solve all their ills by magic…

…but it is the same ’saviour’ that has thousands of missile warheads pointed at them, only 20 minutes from impact at any time.

Round numbers

Finally, one other thing: ’round numbers’. Check out their forecasts:

“The Taiex Index may climb to 8,500″, “The brokerage previously forecast 7,700″

Notice how these are both round numbers, multiples of 100? Isn’t that strange? How do you imagine they got that number and not another number?

a) A complicated computer program crunched numbers and computed statistics, until it produced the prediction: precisely 8500.000! Incredible!

b) Everyone in the office picked a number between 70 and 100. They put the numbers in a hat, took the answer out, and multiplied it by 100.

:-|