|
Take an ant colony as an example, the sheer size, complexity and adaptability of this collective is just plain awe inspiring. Observe an ant as part of a colony, and you will be startled at how intelligently and well coordinated it behaves. By contrast, isolate the ant from its peers and you will quickly notice that it is pretty much useless.
Scientists have only very recently discovered the modalities that make an ant colony or a beehive function the way they do. In the case of ants, it is based on an intricate feedback system that goes something like this: early in the morning a first wave of patrolling ants go on an expedition of sorts. What happens next depends on what they discover. If they are lucky enough to not end up as food, they continue their search until they find something worthy of notifying the colony about. The second wave is on standby during this time and remains so until they have contact with a sufficient number of returning ants. If an insufficient number of ants return or if the frequency of contact with the returning ones is more than 10 seconds apart, the trail is basically discarded in favor of a new one, and the whole cycle is repeated.
This example of what might be described as nature’s version of a mathematical algorithm reveals how the intelligent behavior of the colony is at the collective rather than at the individual level. In effect, the ant does not function in a hierarchical structure with a central command and does’t seem to be aware of the big picture or the ultimate goal. Instead, they follow very basic rules and interact only with their immediate environment.
An ant is only valuable if it is part of a collective just as the usefulness of a communicating device such as a mobile phone is a function of the size of its network of users. It’s at the collective level that an ant colony harnesses it power to behave in an intelligent manner, something it achieves by pooling the actions of its individual members. Swarming produces several advantages from the very obvious, such as confusing a predator chasing a herd, to the less obvious such as identifying the optimal location for building a new beehive.
Swarm theory in finance
Swarm intelligence comes in various shapes and forms. Take the stock market as an example. We all know that the price of a particular stock at any given moment is the result of an aggregation or composite of views of a large number of investors in the market. Its effectiveness depends on the degree of diversity of the participants. A truly diverse group would include analysts, traders, speculators, portfolio managers and other types of investors. The diversity in the backgrounds and objectives of the participants is key because it helps eliminate natural biases.
Analysts, for example, are highly informed regarding the company and industry behind the stock and their decision to invest is based on expectations. Others such as investment managers might be investing for diversification purposes. Yet another group of investors may simply be speculating on the stock. The point is that having a diverse group of investors with differing knowledge and aims leads to stock prices that more accurately reflect reality. This also explains why small stocks with low trading volume and poor analyst coverage tend to have prices that don’t always reflect the fundamentals, providing ample opportunity to generate alpha.
An efficient collective based information system such as the stock market requires that it should be decentralized and the individuals that participate in it should be unbiased or independent from one another when it comes to deciding on an investment opportunity. The idea of pooling together the views of a diverse group of participants is nothing new, however, and has proven to be so effective that at one time even the Pentagon was seriously considering introducing a type of open market auctioning system to help predict future conflicts and their outcomes. The project was axed for security reasons.
Democracy as a concept is another fine example of a form of swarming in practice. Like in the other cases, it is only effective if certain conditions are met. A properly functioning democracy requires at the very minimum a stable environment in which the population enjoys a minimum standard of living. This goes in some ways to explain why democracies tend to be messy in the beginning. In a mature democracy, common sense almost always prevails, but even then we get the occasional hiccup. I guess we humans were probably never hardwired through natural selection for swarming activity. Think of traffic lights as one of many examples of our inability to coordinate without help.
Stock markets and democracies apart, swarm theory is attracting a lot of interest and is increasingly being modeled and put to use in different industries with very promising results.
Link to Lobnek's website
| 
