Skip to main content



Information Cascades and Stocks

The paper I read is titled “Institutional Herding and Information Cascades: Evidence from daily trades”, by Susan Christoffersen and Ya Tang from McGill University. This paper piqued my interest as I am interested in finance and the stock markets, and furthermore the markets are something that many people are familiar with. Seeing the stock charts on Yahoo or Google Finance is something that many people reading this have seen, and applying information cascades to such a widespread topic gives the idea a sense of applicability.

 

The article discusses about specific stocks rather than the market as a whole. As in the Networks Textbook, the authors explain how investors base their buying and selling decisions based on what the rest of the market is doing. To start out, stocks have potential to go up due to change. For example, a company can announce a new program designed to increase revenues. However, the chance that this program occurs is dubious or unsure, and at a base level investors make a bet on the stock off of whether or not the said program will pan out to be true. Similar to the marbles example given in the textbook and in class, a certain investor can begin a cascade by making a decision on how the company will go in the future. Given more investors making similar decisions on the stock, an information cascade can occur. It is important to note that the information cascade forming in this situation is not the exact same as the marble experiment, as every investor has their own semi-informed opinion that is not based purely on probability. However, herding can still occur as often investors may be influenced by previous decisions that they can see in the market and give less weight to their own thoughts. Another way stock herding differs from the marble example is that stocks eventually crash, the growth they experience through herding often breaks. This is because at some point investors can see the true result of a company’s program or strategy, and thus investors become more informed, at least for a short period of time, and get a general sense of whether the stock was inflated too much.

 

It is also important to note that in such a real world situation, certain players can have a bigger influence on herding than the odd investor. If David Einhorn of Greenlight Capital (fun fact David Einhorn is a Cornell alumnus), a billion dollar hedge fund manager, states his opinion on the direction of a stock, that in itself could start an information cascade. In real world situations, some people have a bigger weight than the typical person in influencing others’ decisions. The article also touches on the fact that information cascades for stocks form more often on a daily period as compared to a monthly or yearly period. This is because there is more speculation intraday and decisions are made with less information. This can explain some up and down movement in stock charts. While there may be mini cascades every day concerning a certain stock, over the long run with more information and less speculation (akin to basing a decision off of probability and chance), a stock will generally go in the direction of its actual value. Looking at stock movements through information cascades provides a unique perspective on a major industry.

 

Paper: https://www.mcgill.ca/files/desautels/YaTang-JobMarketPaper.pdf

Comments

Leave a Reply

Blogging Calendar

November 2014
M T W T F S S
 12
3456789
10111213141516
17181920212223
24252627282930

Archives