Efficient Market Hypothesis
is the proposition that all public information relevant to the price of a security is reflected in the market price.It follows that the market price, as a function of time, is a random walk. Sir Maurice George Kendall is famous for painstakingly testing this on stock market data, and finding that, indeed, his samples were random walks. This finding is widely touted as evidence for the Efficient Market Hypothesis. So, how does this go: A implies B; I observe B, therefore A???
See also Mind of the Market.
Expertise
At the height of Artificial Intelligence of the 1980s "expert systems" were in the centre of attention. The recipe was simple: the "knowledge engineer" interviews an expert (say, a specialist in internal medicine) and expresses the elicited "knowledge" as rules for processing by the expert system being built. It is useful to remember why it didn't work: the experts did not know why they did what they did.This is more important than just an episode in a fad of the past. Without it one can debate endlessly, and fruitlessly, about the relative merits of rationality and intuition. In the case of expert systems disciplined application of rationality failed. The subsequent success of machine learning techniques can be interpreted as a way to find an effective role for intuition. In machine learning, one merely records the expert's actions, decisions, or judgments without expecting any reason. Machine learning processes large numbers of associations between situations and actions in a way that turns out to reproduce or improve the expert's performance.
These findings support and supplement John Kay's current book "Obliquity" and his take on Malcolm Gladwell's "Blink".
Incentive Compatibility
Invisible Hand
How is it that a great many economic agents (producers, consumers),each pursuing its own interest jointly effect an outcome that may not be all bad, at least compared to Soviet-style centrally planned economies?
Adam Smith (reacting to a different example of state direction of the economy) answered with this description of a typical economic agent:
... he intends only his own gain, and he is in this, as in many other cases, led by an invisible hand to promote an end which was no part of his intention. Nor is it always the worse for the society that it was not part of it. By pursuing his own interest he frequently promotes that of the society more effectually than when he really intends to promote it. I have never known much good done by those who affected to trade for the public good. It is an affectation, indeed, not very common among merchants, and very few words need be employed in dissuading them from it.The metaphor of the Invisible Hand is amazingly powerful. It was invoked for a hundred years before Walras managed to give meaning to it, which was in the form of a system of equations. Then half a century went by before somebody investigated whether such systems had a solution. In the 1950s the subject was closed with the Arrow-Debreu theorem proving the existence of a solution and a proof that it represented a unique and optimum equilibrium. The fact that the theorem was proved under extremely restrictive conditions that are unlikely to prevail in any practical situation has not diminished the hypnotic power of the metaphor and its claimed implication that free markets lead to the best of possible worlds (with everything in it a necessarily evil?).
The metaphor provides insight without necessarily being applicable. It is invoked by conventional academic thinking in economics as a blanket justification of a certain style of government that serves vested interests. There is hope coming from a new orientation in academic economics exemplified by the use of game theory for mechanism design and studying actual human behaviour.
