By Doug Tucker
It is generally accepted that there is a relationship, or a measurable correlation between certain markets. There have been many articles, books, trading systems, and approaches based on intermarket relationships. If you turn on the financial news you will usually hear an explanation such as: the stock market was down today because crude oil was up, or maybe the stock market was up today because crude oil was up. It can even get more complex. You might hear something such as the stock market was down today because crude oil, being up sharply today, caused the bond market to be down. With the bond market down and interest rates up, the stock market was under pressure and down it went. Etc. Etc.
I find it a good idea to question everything, especially commonly held beliefs that might not hold up under close inspection.
One can very quickly and easily test these tried and true intermarket relationships with easily available free data and a simple spreadsheet or charting program. The quickest function to use is the simple correlation study, where you plug in one variable, and then compare it to a second variable. The study tries to find a correlation between two data series. If it finds a positive correlation the value will go as high as + 1.0. That would be a perfect and positive correlation between the two series of prices. If it finds a perfect correlation, but inverse, or negative then it will have a value as low as - 1.0. Readings near the zero line would show no discernible correlation between the two samples.
It is rare to have a perfect correlation between any two market for a very long period of time, but most analysts would probably agree that any reading sustained over the +0.7 or under the –0.7 level (which would equate to approximately a 70% correlation) would be statistically significant. Also, if the correlation value went from a positive to a negative correlation frequently, the relationship would most likely be unstable, and probably useless for trading.
If a picture is worth a thousand words, I'll save some of mine by directing you to my blog where I have many examples. I've included a couple of similar markets that did show a high level of correlation just to show what a good, stable correlation would look like. The rest of the markets shown are of relationships that most people believe to have stood the test of time, but the charts show otherwise. The address is below:
http://tuckerreport.com/articles/intermarket-analysis/
The most widely accepted correlation for as long as I can remember is the inverse correlation between stock prices and interest rates. It is assumed and generally accepted that as interest rate go higher, that stock prices go lower, and conversely, as interest rates go down, then stock should go up. On the second point, the Japanese market was in a huge downtrend for a time, and lowering interest rates to near zero did little to help. Back to the first point, the interest rates recently have been increasing steadily and many stock markets around the world, including the Dow Jones Industrials, have recently made all time highs.
In addition, crude oil is supposed to be bearish for bonds and stocks, yet crude is near an all time high along with stocks. Gold is supposed to be inverse to the stock market. It has been briefly a few times in the last couple of decades. Over the last few years it has been in an even steeper uptrend than stocks. As this is being written, the stock market has had a huge sell off, and gold came right down with it. Looks like the inverse correlation has been divorced, or at least separated.
What does any of this tell us? Is the much-touted intermarket analysis a waste of time? In a word: yes.
Many trading systems were bought and sold based on inter-market relationships. Every one of them now is most likely worthless. The problem with those systems, and intermarket approaches in general, is they were designed on relatively small samples of data. If they used much larger samples of data, they most likely would not find the intermarket relationships, they thought existed, to be stable and robust. When an intermarket relationship becomes obvious to everyone, it’s on its way to becoming useless. What everyone knows isn't information that helps make successful trading decisions.
Even if you could find a long lasting, intermarket relationship, how would you use it? I assume the purpose of that analysis would be to find an edge by figuring out which market was leading the other. How can you tell which is leading, if indeed one does lead. You can try to guess at the fundamentals, but that would be a guess about the future. Why not just trade the market you are guessing about and not compound the situation by assuming the guess, if correct, will affect the other market in the way desired. You might be right in your guess, but wrong by assuming the relationship between markets will play out the way you anticipated.
There is even a long held belief that stock market averages influence most of the direction of individual stocks, some say by 70%. But obviously, there are many instances of stocks going down in up markets, and stocks going up in down markets.
My conclusion is that each market should be charted and analyzed individually. Intermarket relationships are fun for discussion, and for financial journalist, but I don't believe they can help in making trading decisions.
Doug Tucker has a blog with daily commentary on stock indexes, precious metals, and other markets. There are many articles on technical analysis and indicator design and interpretation. To visit go to: http://tuckerreport.com/
Article Source: http://EzineArticles.com/?expert=Doug_Tucker
http://EzineArticles.com/?Intermarket-Analysis---Do-Intermarket-Relationships-Help-Make-Trading-Decisions?&id=662428
No comments:
Post a Comment