Beware of the ‘fallacy of composition’

[plusnq]

To be honest I don't think you properly understand Taleb - you should probably give it a second read.

The turkey example is not about tail risk - it simply highlights that the past is not a good indicator of the future. Take a forecast of oil price (based on 20 years of historical trend data) prepared in 2001 for 2008. It will be so far off the mark it is ridiculous. It's not even in the tail! Well, technically in a normal distribution everything is in the tail as it never hits 0 but you know what I mean - it would be a 1 in 1,000,000,000,000 event.

The connection Taleb is making is this: if time series data is useless, then the variation around the mean is also useless, and therefore our assessment of the probability of a tail event is also useless. The losses the investment banks just took were not even in the tail of the risk managers working at these firms - they were mathematically near "impossible" according to their historical distributions of events around the mean.

So the point is people who get comfort from 30 years of property history could lose their head! People need to be more critical and think it through.

edit: there is the possiblity that I am the one who doesn't fully understand Taleb - people take out of it what they want to take out of it I guess.

yM

Would that variation from the mean include property prices? So taking Taleb's view (as expressed by yourself) then property prices that have deviated so far from the mean may be a continuing event as our ability to predict the correct price is not possible with the data series available. Perhaps you are right. Some

People need to be more critical and think it through..

The ability to think critically is the essence of any research. It includes not only the ability to examine a proposition from many angles but also the ability to allow oneself the luxury of changing your mind when the facts change.

Cheers

Shane

 

[Andrew_A]

Matusik has noted this for the Brisbane market, a lot of investing stock coming onto the market and hence putting downward pressure on 'median' sale prices. I think he mentioned an issue relating to super rules and older investors selling, though someone else might elaborate on that.

Anna,

I agree with you on this one. Went to inspect a property this morning that was described as "seen better days" - and I agree, possibly 30 years ago.

What I'm noticing is that a lot of the properties I am seeing are currently rented out. So I suspect it's a lot of investors heading for the exit, either because they fear an impending crash, or not keeping up with payments.

First the downturn in median prices will reflect the low value PI properties, then depending on the impact of the US recession in Australia, it could be the higher quality OO properties coming on. We may see a lot of spikes and dips in the median house price over the next couple years.

 

[BayView]

Look if you are worried about such a scenario, don't consult your favourite gold bug site or US based housing tale, talk to investors from the Gold Coast circa late 80's who had to sell after buying stuff OTP, people stuck with two tiered stuff and most everything in the 90's. That will give you a much better baseline for some scenario calculation imo.

All that type of investing is really only speculating, or they are acting on the OPINION of someone else.

If people really did their due diligence THEMSELVES, they would know that the property they had bought was overpriced.

It was the same with the apartment glut in Melb a few years ago.

There were loads of OTP purchases by cashed-up professionals who wanted to follow the Jones, and could afford the ride, but in typical fashion they were simply armchair investors and got burned for their ignorance.

The bottom line is; don't sell in a market slow-down unless you have to, and if you have to so be it. Take your medicine.

Always use risk management strategies such as lower LVR's, lower repayments as a % of income, strong rental demand areas (not in high-rise apartment blocks where everyone else is doing the same), depreciation etc and you should never have to sell until it suits you, or, if you do have to sell during a downturn, you will be able to minimise your losses.

 

[Andrew_A]

gobble gobble / chop

Rule 7 - The greatest inferential mistake: this event never happens in my market. Most of what never happened before in one market has happened in another. The fact that someone never died before does not make him immortal. (Learned name: Hume’s problem of induction).

Something I found on the web.

 

[keithj]

I have read Fooled twice from cover to cover, done some brief reviews for my blog and refer back to it periodically for my own personal humiliation, I have read Black Swan once (tedious book), all of the material from Taleb I can source on the net and magazines (He's in quite a few trading mags) including his blog and web page and recorded lectures of which there are many, interviews and so on. Also I have researched Taleb's own trading record and some very interesting opinions from people smarter than me about his own performance with 'Empirica'. So it's not a question of effort on my part but rather a question of comprehension.

YM,

Andrew is v. politely saying that .....

there is the possiblity that I am the one who doesn't fully understand Taleb

 

[Andrew_A]

Keith I'm really a beginner (but learning) in the area of probability, statistics and the financial markets, Taleb is an options expert and I can't begin to understand his more complex ideas, though Fooled by Randomness does an excellent job of explaining his world view with no technical jargon involved.

It's reported he is charging 60k per speaking engagement presently which is an interesting thing to consider.

This link has an interesting view about NT. Basically I have seen it said that NT's thoughts can be summed up as 'S*** happens', which is hardly going to help you make money as an investor.

** edit.. Also one of my favourite trading blogs at the moment has several good articles on NT.

http://falkenblog.blogspot.com/2008/03/taleb-on-bloomberg.html

 

[keithj]

This link has an interesting view about NT. Basically I have seen it said that NT's thoughts can be summed up as 'S*** happens', which is hardly going to help you make money as an investor.

Interesting link. And I'd agree that 's**t happens' is a good synopsis. His view that history is useless doesn't help the average investor. History is based on the effect of the sentiment contained in 6 billion brains, and those brains tend to change their minds fairly slowly. I'd say that we can expect the future to look fairly similar to the past. However, we will experience many 1 in 100yr long tail events (tsunamis, 9/11s, iraq invasions, oil shocks, baby boomer retiring and many others). I still say diversification is the free lunch that will mean that I can still invest with high SANF although I am absolutely sure that s**t will happen.

 

[Pete]

In summary, I just want to caution property investors on this forum from thinking that if they do have to sell one or more of their properties it will probably be, at worst, only a small discount from the current price. There is a significant risk that if a lot of property investors decide to sell one or more of their investment properties at the same time the price they receive for these properties may be significantly lower than they expected.

Thanks for starting the thread, Scallywag. The discussion seems to be wandering away from your comments. For sure that scenario has played itself out in the US & UK recently.

There are always risks. What you have described will trouble (bankrupt) a few property investors one day. It is essential to be aware of risks and have strategies to deal with them. Investors who ignore these risks are gamblers and have much greater chances of failure.

Thanks again. Regards,

 

[yieldmatters]

This link has an interesting view about NT. Basically I have seen it said that NT's thoughts can be summed up as 'S*** happens', which is hardly going to help you make money as an investor.

Thanks - good read.

Interesting he wrote that before the credit crisis got into full swing ...

Eric Falkenstein - am 2007-06-20

And said the following ...

And of course he was dead wrong on VAR, in that in spite of his criticisms it became ubiquitous as a framework for amalgamating risks from different instruments into a single metric.

Since then the investment banks have annoiunced billions of dollars in losses and VAR has been badly damaged as a methodology (that is what the banks used). I wonder if he still holds that view? I think VAR has it's place but it needs to be mixed with common sense. Again I go back to my oil price example - VAR would never have picked up this risk - well it would have but it would have been a very, very, very low proabability.

 

[plusnq]

Thanks - good read.

Interesting he wrote that before the credit crisis got into full swing ...


And said the following ...


Since then the investment banks have annoiunced billions of dollars in losses and VAR has been badly damaged as a methodology (that is what the banks used). I wonder if he still holds that view? I think VAR has it's place but it needs to be mixed with common sense. Again I go back to my oil price example - VAR would never have picked up this risk - well it would have but it would have been a very, very, very low proabability.

YM

The problem I have with this idea of your oil example is the fact that the price is marked to market every day. So inputting different prices will give different values for risk and the prudent fund manager or private investor will take into account the changing risk and modify their asset allocations accordingly. The investment allocations aren't fixed in stone.

Andrew has said he sold some property in 2007 to manage his risk and that is fine. Others who may have been less leveraged may not sell as their exposure may be less. It doesn't matter what the pholiosophy of an investor is, the prudent allocation of funds in a diversified portfolio is the only free lunch in town.

When a person decides to become an investor they have to choose what investment they can live with (their SANF) due to their personal circumstances, biases and market conditions. It is often not the best choice (usually due to their limited funds available) but they are at the start of a long learning curve that hopefully will be profitable due to their persistence and desire to learn. If they are lucky then they may make some good money early. If not then hopefully they will evaluate what happened and learn from it. It is at this point that most newbie investors fail and leave.

Experienced investors will watch the markets, both micro and macro and allocate their investment resources according to their skill set, experience and beliefs. They would input rising oil into their investment models and make adjustments for that.

In the example of property they may decide (or not) that allocation of resources closer to the city centre and amentities with good transport links and hubs is the correct allocation of their resources. So they may sell some of their property that doesn't fulfill that criteria and make an appropriate reallocation of funds. Only time will tell if their allocation was good or bad for their portfolio. This is in part why diversfication works. It mitigates some of the risk.

Good luck with your investing.

Cheers

Shane