When I was reading the technical analysis, I was thinking why it might have some reasonable ground. When I was watching the candlestick charts, it indeed looks like there is a trend (long-term trend, short-trend ...). Well I reminded myself the things that the market efficiency believers used to teach me: even a Brownian motion can look like it's not stochastic. But ... is there seriously a reason for the trend?
There may be one. Trough trading, I find one thing that many academic papers ignore: people (especially, the traders) have limited energy and time to pay attention. Although the market has already had so many financial professionals, they are still not enough to absorb so much information, even with the support of back desk of researchers. The information's spread, especially for the areas not many people are paying much attention, is not always very fast. It can take one day, sometimes two days, sometimes even longer. Take the most recent example of TVIX and GAZ/UNG: the premium was so high, and it was not like all the people suddenly find that -- some people posted this on StockTwits weeks ago, but nobody was paying attention. As more and more people realize, they followed into the arbitrage.
Economics news or big issues publicly known are quickly reflected in the market prices, but not all the information has the priority to be known. "Economic Calendar" in WSJ posts the big information releases, but among tens of thousands of pieces released everyday, not all of them are noticed in the real time.
For commodities and government bonds, there are relatively limited variations. Copper is copper, gold is gold, information is not that heterogeneous compared with stocks (commodity traders, pls correct me if I'm wrong). But for stocks, every listed company is different, and every one is complicated inside. Aside from the quarterly/annual reports you hardly know what is happening now. Unless the new information is published, no one knows the everyday dynamics and if you don't find new information about a company, of course your attention moves away to others.
But what's the fluctuations around the main trend? It's because the informed don't know whether they are the last or the first to know in the market. If they think they are the last to know a positive news, they should sell off in the very short run; if they think they are the first, they hold. But people don't know, so they have uncertainty, then the "stochastic" things begin to work, and a lot of research were discussing whether the weather/temperature/balabala statistically affects traders' sentiment.
Then I have a research idea: How can I test the speed of the information spread? Make a rumor and see how long it affects the market price? (experimental economics !) Well, no mention if it's illegal, it can be affected by a lot of random effects like other big news or something. Though I haven't figured out a good way to research it, I guess it's an interesting one.
No comments:
Post a Comment