I know people who go all in on one philosophy. They’ll either only look at the fundamentals, or only use technical analysis. Honestly, this is not healthy. Investor sentiment drives the market. And investor sentiment, barring any external factors, follows certain patterns, which is why technical analysis is so useful. But one would have to be a fool to ignore external factors.
I am all about theory, but theory drives practice. It can help answer the question of when to rely on technical analysis, and when to avoid it.
A Simple Model of a Stock Market
Many traders and investors rely on technical analysis. I personally rely on both Western line and Eastern candlestick analysis. However, that does not mean that it is all I use. Determining when it is appropriate to rely on technical analysis, as opposed to other sources of information is difficult. One clue for the limits of technical analysis is found in a study by Caldarelli, Marsili and Zhang.
In 1999, the three worked on a model of the stock market. The model used basic investor strategies tied to human psychology. They ignored external factors, such as rumors and news. Even with this simplifying assumption, there was a significant amount of similarity between the way the real markets react and the way the model functioned. Unfortunately, the authors use a poor word choice, saying that the model “behaves…realistically.”
It does not mean that all of the market’s dynamics can be determined from internal mechanisms. From the realization in figure one of the study, we see that there is a lack of overall trend, and the price fluctuations are absurd. At one point, the price for the security was over 900. At another, it was down to 25. Then later on, it was back up to 500. This kind of pattern occurred repeatedly, while the overall price never trended anywhere.
This makes sense, as there is an assumption that the investor expectations are neither overall positive or negative, and there is no overall shift in investor expectations. Clearly if stocks worked like that, no one would be investing. If it were normal for stocks to be able to loses 97% of their value at some points in time and yet also balloon 20 fold or more at others, it would be kind of absurd.
So in at least some respects, the model does not behave realistically. Essentially, the model is a zero sum game model. However, the markets are not zero sum games, at least not when there is a significant level of investor activity. However, it does behave realistically in another way: locally. The local dynamics are determined by human psychology and price history.
But overall trends are determined by something else. Stocks trend higher over time, in part, because the underlying asset-the corporation-is growing in real value. The stock market as a whole trends higher because the economy as a whole grows. But of course there are cycles of growth and decline, and that has an impact on the global dynamics of the market. And the authors do admit that internal factors alone are not enough.
It is clear that by no means one can conclude that our model captures all the relevant aspects of a real market. As already said it misses the effects of external drive. More importantly, it does not contain adaptive dynamics of the player’s strategies. Further studies should answer the question: what are the essential elements in a model that will reproduce realistic results? In this perspective our model can be considered as a first step in this challenging direction.
Essentially all of the article is theoretical. However, I did mention that theory leads to practice. In that regard, I want to end on a comment about the current state of the markets. I do not like to give precise advice or make exact predictions. But I will offer a word of caution. Someone who is relying purely on technical analysis might be in for a shock. Right now, stock markets are sitting at incredibly high valuations. Explaining why is beyond the scope of this article, so instead I will offer a link to a prior article on why I am concerned about the markets.