Forget Data Minin U
Post on: 29 Апрель, 2015 No Comment
by Alexander Green. Chief Investment Strategist, The Oxford Club Monday, March 15, 2010 Wisdom of Wealth
by Alexander Green. Chief Investment Strategist
Monday, March 15, 2010: Issue #1216
There are lots of theories about how to predict future stock market performance.
Most of them are the sheerest nonsense.
There is, in fact, only one indicator that virtually guarantees you will be on the right side of the market.
But first, let’s look at the most popular source of misconceived market predictions: Data Mining.
Data-Mining Investing Strategies. Don’t Expect Reliable Answers
Every year, investors pour billions of dollars into data-mining investing strategies. They come with impressive-sounding features like proprietary trading tools or McMillan Oscillators. But the money might just as well be invested on a coin toss because the reasoning behind it is completely baseless.
What dating-mining strategists typically do is back-test, or simulate what would have happened if you’d used a particular technique in the past (typically without incurring any fees, trading costs or taxes).
It’s not hard to look back and see what the market did under certain circumstances. For example, what has the S&P 500 historically done when the Federal Reserve starts raising interest rates.
In truth, it doesn’t matter, regardless of what parameters you use. You can look at GDP growth, inflation levels, stock market valuations, long-term interest rates, the price of gold, or the value of the Swiss franc to the dollar.
As irresistible as data-mined numbers sound for making predictions, the whole shebang is based on a faulty premise.
Don’t Fall Into the Data-Mining Trap
The first thing you learn in Statistics 101 is that a positive (or negative) correlation does not infer causality .
In other words, even if stocks have rallied every second Thursday in May for the past 50 years, it doesn’t mean they will rally this year on the second Thursday in May. Data-miners regularly turn up meaningless correlations and claim they have discovered how to divine the stock market.
If history could determine what the stock market is going to do next, the world’s richest investors would be historians, data-processors and librarians. And that’s not the case.
The hypothetical results of data-mining always crumple when they collide with real-world investing.
There is only one sure-fire market timing strategy. It’s only available occasionally. Plus, it takes guts and a genuine contrarian spirit.
The Best Time to Invest in the Stock Market
The best time to invest is when you see extreme valuations coincide with extreme sentiment.
For example, if unbridled optimism reigns supreme — as it did during the tech-stock bubble ten years ago and at the top of the real estate craze four years ago — and prices are sky-high, you can bet your last dollar that prices will soon come tumbling down.
By the same token, if abject pessimism is the order of the day — as it was during the financial crisis one year ago — in concert with rock-bottom valuations, you can bet that a rally isn’t too far distant.
Indeed, the Dow is 62% higher today than it was a year ago.
That’s why you might want to invest a few dollars today in one of the world’s cheapest and most unloved markets: Japan .
What’s the difference between data-mining and seeking out extremes in sentiment and valuation?
Two things, really. The latter strategy doesn’t require a rabbit’s foot — and it works.
