Statistics, The Funds, Booms and Busts

Firstly: I recently had a comment about high-performing funds.

The short answer: some people get lucky. Nobody hears about the underperforming funds. If you take a consistent definition of "a fund" -- and track them over time without changing it -- and do a real average, then you might get a number you can believe. What's more likely is that some statistically unwise reporter has found 5 or 10 funds which have all been profitable, averaged their results, and gotten an above-average return.

Simple statistics prepared by anyone other than an expert, with anything less than a full understanding of the statistical methods used to arrive at them, do not carry enough meaning to do be certain about their implications.

It's equally possible, for example, that if you take the last three boom years, the average fund return *was* around 17% -- just as it likely was for many individual highly leveraged traders. However, if you extend your period out to the long term (say 15 years) to incorporate some crash periods, the figures will look less rosy.

By all means, consider the value of a fund, but don't place infinite faith in them.

The ASX has recently had its single biggest drop since S11, and has been falling steadily for several days. Yet, considered over only a 1-year period, returns are still in excess of the 10% rule-of-thumb figure mentioned earlier. While there has been a recent plunge, investors are still experiencing long-term rewards. The intensity of the current fall is high, but its duration has been short, and has not had a terrible impact considered over the sufficiently long term. The longer the period of time under consideration, the lesser the impact of extreme events if they can be survived.

To speculate, one reason for the size of the current fall might be that individuals are taking profits. Relatively few people, perhaps, have bought shares at recent high prices, while relatively more, perhaps, are realising 25% profits from purchases made several years ago. Considered against yesterday's prices, there has been a big fall, but many traders will be considering not yesterday's price, but their purchase price. Since traders may own shares for many years, medium-term price history is highly relevant.

While in some senses this seems ridiculous (why would anyone not want to sell at the high, and not be dismayed by the drop), it also makes sense that individuals will be making decisions against profit targets, trading overheads and their own particular strategies. It seems like the better risk to lock in a 25% profit than hope for a return to previously high values to gain 35%. Risk-averse traders would prefer to protect their returns.

Cheers,
-MP
(This is not investment advice. I'm just this guy, you know?)

Free ASX OHLCV end-of-day data

http://www.ipo-australia.com/

Has free pricing data available for download. Not quite the realtime tick-data I'd ideally like, but not bad for the price!

Cheers,
-MP

Quality of Information

One of the things which I have found challenging is sorting through the abundance of information and misinformation that is available.

Let me state from the outset -- I do not subscribe to any form of supernatural intervention in investing. As far as I am concerned, it is a strictly human affair.

That cuts out a frightening amount of the available literature at the get-go. What remains breaks up into "Stuff Based on 1950s Research", "Fluffy Contentless Books for Beginners", "Ludicrously Complex Books", "Sensible Tips" and "Miscellaneous".

What's surprising to me is the lack of anything to bridge the gap between sensible tips and a useable guide to investing.

I think I have identified the reason -- investors are not hackers. If investors were hackers, every Joe Soap would have published information on the web for criticism and shared knowledge. There would be a lot of crap, but there would at least exist *some* more complete documents describing how to do that trading thing.

This is corroborated by the lack of open-source software for trading -- or software which even *runs* on linux. Trading has not yet been hacked.

Partly, this must be because hackers don't find trading interesting. However, I'm surprised, because at some level the information is readily available if you can put the time in to look for it, and because it involves improving a system.

Anyway, this blog started being about quality of information. I really, really recommend spending a lot of *effort* and not a lot of *money* on learning about shares. Very few books will directly or completely address your needs.

Over the coming posts, I will attempt to describe some of the better sources of information which I find. Book reviews are one way in which this will happen, but I will also attempt to review ideas, websites and organisations along the way. This lends itself to wikification, but I don't have the time right now.

In the meantime, I'm starting with baby steps. I'm risking amounts of money that I can afford to lose, and spending a lot of time at the research level.

(This is not investment advice. I'm just this guy, you know?)

Cheers,
-MP

Beating 10%

MelbournePhilosopher

Investing in the sharemarket is a game of trying to beat 10%. Historically, this has been the long-term return on Australian share investments. Bearish markets, bullish ones, recessions, depressions booms and busts simply vanish into the statistics over a long enough timescale. If you use a risk-minimisation strategy, and stick it out for long enough, you could reasonably expect a 10% return. (Source : http://www.asx.com.au/investor/pdf/getting_started_in_shares.pdf -- see the graph of expected returns). This is a compounding figure, so your absolute gains increase over time as you get interest-on-the-interest from previous years.

So, are you feeling lucky?

If you just buy one share, you're at the other end of the spectrum. Your money will experience far greater variance over time, as a single company is far more exposed to both danger and opportunity than the market as a whole.

Corrolory: If you buy a specific company, you are in fact using your money to wager that the returns will be greater than 10%. You are gambling with your money, risking a worse return in order to gain a better one.

If you don't have any reason to think a company will outperform the average significantly, don't buy it. If you want to get better than a 10% return, do a lot of research first.

Another way of thinking of this is in terms of the Red Queen problem. The Red Queen from Alice in Wonderland is forced to run as fast as she can just to stay in the same spot. The sharemarket is the same -- as soon as you're in, you're in a marathon race where in order to stay with the crowd you have to run. If you pace yourself, you should be able to finish somewhere in the middle. The pack is running at an average of 10% p.a. over the long term.

Things are, of course, a lot more complicated than this. There is more than one "rate" which you should be trying to beat. One is inflation, for example. If you are making less that inflation, then you really *are* going backwards as your money loses buying power. If you are making 6%, you are merely less ahead. However, it is the 10% rate against which one should primarily be assessing performance.

The basic reason for this is that there exist good strategies which make the 10% rate a very plausible target for any investor.

All share purchases are either attempts to reduce risk and thus achieve the 10% average, or to increase risk and beat the 10% average.

Cheers,
-MP

(This is not financial advice. I'm just this guy, you know?)