[MUSIC] Hi, I'm Tony. >> Hi I'm Olivier, so Tony what are we going to talk about today? >> Well today, we're going to introduce the concept of market efficiency, and we're mainly going to answer three questions. How is information integrated into prices? Can we categorize the different forms of information? And what does all this imply for the performance of various trading strategy? >> So if I summarize, what we're try to learn is how to be super efficient in efficient markets. So I know that our students are super efficient. So now how do we define efficiency of markets or market efficiency? So in fact we can summarize that very easily, is that prices should reflect all available information. So now this is in fact a summary, but which is not really former. So how can we define that in a more formal way? And to do that we can go to a very old paper by Eugine Farma and he received the Nobel prize for that paper and that paper was published in 65. So what he did is to characterize the efficiency of markets depending on the information structure that you have and how is it reflected in the prices? And so what do we have if you look at the slides, so we have what is called weak efficiency, semi strong efficiency and strong efficiency. So the first one is the weakest form in the sense that the prices only reflect the information contained in past prices. So if we go to the semi strong efficiency it's a little bit indeed stronger in the sense that it should also reflect all publicly information like for example, what you can read in the newspapers. And the last one, which is the strongest one form of efficiency. The prices should also reflect all private information. So for example, the information that insiders or CEOs that said, their heart can have about the performance of their company. So I have just talk about the concept introduced by Eugene Fama but that concept of market efficiency is in fact much older. So if you look at the first paper, not coining the term but at least introducing the fact that prices fluctuate randomly It dates back to Louis Bachelier in 1900, so it was very, very early. Bachelier in fact did in his PhD thesis in mathematic, if you look at the introduction of this very old thesis, he was already quoting the quote that you can read on the slides. Basically, what you are telling us is that indeed prices should reflect all available information and should indeed fluctuate randomly. So this is as you can see a very old concept and it was relatively the same time that people like Einstein, Vienna or Brown introduces what is called now the Brownian Motion of the Vienna process. Which is used now a days in order to mother the evolution of particles in physics. So indeed if you look at the particle in physics we know that, it fluctuate very randomly. Now that old work of Bachelier was completely forgotten for a number of decades. It was only found back by Samuelson, so another Nobel prize in economics and it was in fact published the first time in English by Kutner in 64. So now, let us look at the second very important piece of work and it's again an old paper by Cowles. So Alfred Cowles was a benefactor of the econometric society. And he was also a founder of what is called nowadays Cowles Foundation, and the Cowles Foundation is still nowadays hosted by Yale University. And what did Cowles is to make a very interesting study about the recommendation of investment professional and what he found is basically that there is nothing, nothing in the sand that professional infer can not all guess what the market will do. So this is the work by Cowles. So the third piece of empirical work and again quite old paper is work by Working Cowles and Jones and what they did was to find exactly the same findings as what Bachelier did for commodity prices. So what they found is if you look at time series of economic data or time series of equity pricings again that fluctuate randomly. So now in fact Tony, we'll summarize a little bit what we have learned today. >> So first of all, one important aspect of this discussion is that there are indeed three separate forms of market efficiency. They can be defined and characterized by the type of information structure that they refer to. As the very explicit introduction that Olivier gave us, we know that this is a rather old concept but it is still very important today. Before continuing with our discussion of market efficiency, we would like to ask you a very simple question. This will be a poll that's played on your screen. If markets are more efficient, do you think that this implies that prices are more or less predictable? [MUSIC]