Trend Analysis

When you do analysis, depending on the data used the results have some degree of objectivity. For example, if you have a population of rats and feed them large quantities of a substance and have another population that has the same diet except for the particular substance, variation between the two populations may be related to the presence or absence of that substance. The more you can eliminate other potential causes of variation, such as genetics and environment, the results increase in value. However, even in a controlled experiment, the potential for some unknown factor to enter the experiment that normally you need to prove the results are repeatable, meaning that someone performing he same analysis halfway around the world would get the same results.

Now when you have analysis performed in an area where variables cannot be controlled the reliability of the results is questionable. For example, you often hear free market advocates talk about how a free market is the best way to promote growth and prosperity. Of course, it is hard to prove this experimentally so they use logic and examples. Now, the United States with something of a free market economy did better than the Soviet Union under communism. This may have been because of the economic systems or any of a million other differences between those two countries. Certainly I have no way of proving the free market economy of the United States was not a significant influence in the outcome, I just can't prove it one way or the other.

Now, consider some of the analysis you hear about financial trends. There are so many different factors influencing most industries that even if you successfully spot a trend, it is impossible to know for certain that there is not some countervailing trend you haven't spotted. Of course, if you can accurately predict that smart phones are taking more and more market share and therefore manufacturers of smart phones and smart phone components should do well, you also have to factor in how much they are losing if they currently manufacture dumb phones and if the phone they come out with is going to be popular or successfully promoted. The i-Phone was a phenomenal success and the Palm Pre and Palm Pixi weren't. I have read many reviews where the Palm products were considered technically better than the Apple phone but one took off with the right amount of hype and growing market share and one didn't.

Of course after the fact you see analysis as to what happened, and sometimes this post mortem seems perfectly logical, however, that is after the fact and somewhat irrelevant to most investors. Consider the events that happened today and the impact they had on the market. First China's announcement concerning the Yuan led to a higher start as this was perceived as another indication that the economic recovery was continuing and that foreign products would become more affordable in China. This trend held the market up for a while and then we had the assault on the banks as first one analyst said that a double dip in the housing market was a certainty and another analyst said that the financial reform bill would be a disaster for the banks. Further, we found out China was holding more gold than previously thought. Each of these factoids and analysis had some impact on the market and by late afternoon, the earlier increases had been erased and the overall market was slightly down. Of course the afternoon sell off may have nothing to do with the negative comments, it could simply be a combination of fund managers taking some profits and others seeing a trend develop, jumping aboard.

In fact, the one thing that is certain is that no trend goes unobserved. Once a sell off becomes clear, selling becomes the thing to do. You can always buy the stock back cheaper after it goes down some. If you are shorting the market, a sell off is a cause for jubilation. In fact, one thing you can count on is that on days with late day drops, there will be a bounce near the close as some traders get out of positions to avoid overnight risk.

Of course that trend analysis has very little real validity.