Hacker News Reader

Comment by lisper | Hacker News Reader

For any big market movement you will always be able to find the person who acted on it first. That person will always look like a prescient genius if you just look at that one data point. To show a counterexample to the EMH you would need to exhibit someone who can consistently beat the market at odds better than chance under proper controlled conditions. No single data point can ever do it.

View on HN

> To show a counterexample to the EMH you would need to exhibit someone who can consistently beat the market at odds better than chance under proper controlled conditions.

See perhaps Renaissance's Medallion Fund:

> The Medallion fund is considered to be one of the most successful hedge funds ever. It has averaged a 71.8% annual return, before fees, from 1994 through mid-2014.[32]

* https://en.wikipedia.org/wiki/Renaissance_Technologies#Medal...

People speculate that while mathematically half of trades must win and the other must lose (for each side of a buy-sell), that Medallion may perhaps 'win' something like 52% of trades, but their volume is so high that the (e.g.) 2% gives them their profit just like The House only needs a slight edge when it comes to casinos.

Figuring out whether a given performance level is statistically significant is incredibly difficult. You have to take into account a myriad factors: how many other funds were there? The more total funds there are, the more likely you are to find outliers. What was the risk-adjusted return, and what were the underlying market conditions? It's trivial to take a consistently rising market and amplify its returns: just apply leverage.

The fact that Medallion had this extraordinary return in a period that included the 2008 crash is remarkable, but even then you would only have had to predict a single black swan in order to avoid being impacted by it. And even I saw that one coming back in the day, and I don't really pay that much attention.

The real problem with the EMH is most people consistently do worse than the market not better. In a completely efficient market every trade should be equivalent, but when you look at what happens when millions of the uniformed try and day trade they on average lose massive amounts of money.

PS: I have tried to time the market exactly 4 times and it’s worked every time. That does not mean the EMH is dead, that’s just like how I am actually up on slot machines and have decided never to play again.

Right. To show the EMH is dead you have to show that someone can beat the market with odds better than chance under properly controlled conditions. In a market with millions of participants, you would expect there to be a few who beat the market purely by luck many times in a row. In fact, in a population of a mere one million people, you would expect to find one person who beat the market 20 times in a row without a single miss by pure luck! Four for four is not even remotely remarkable. One person in every 16 will be able to do that.

Sure, that's all correct. But before showing it is dead it would be useful to have empirical evidence that it was ever alive. And the evidence is very mixed at best. The history of the EMH is a battle between those showing where it breaks from reality, followed by increasingly desperate attempts to save it with some additional explanatory factors.

It's a nice model. But like most economic and financial models shouldn't be held as gospel.

That's like the theory of energy. Energy can't be created or destroyed, until someone does and whatever they did gets merged into the definition of energy.

The EMH doesn't say anything about anyone not being able to "beat" the market. Only that prices move until they reflect the available information. While prices move, there can be winners.

Also, information is not uniformly available in real markets. That subset of the traders who have all the information will beat those who have only some of the information.

There is a lot left out of the hypothesis. For instance, even if every trading node in the system has access to all the information, do they all have access to every pieces of new information at the same time? If some of them get it before others, then they have an advantage. If all nodes eventually have the information, then the prices will settle according to the hypothesis when new information stops being available.

Worth pointing to this classic: “The Super-investors of Graham and Doddsville” [0] by Warren Buffett.

[0] http://csinvesting.org/wp-content/uploads/2014/10/The-Superi...

Two points about Buffett:

* over the last few years (15?), he's actually trailed the S&P 500 (IIRC)

* 2013 paper "Buffett's Alpha"

> Berkshire Hathaway has realized a Sharpe ratio of 0.76, higher than any other stock or mutual fund with a history of more than 30 years, and Berkshire has a significant alpha to traditional risk factors. However, we find that the alpha becomes insignificant when controlling for exposures to Betting-Against-Beta and Quality-Minus-Junk factors. Further, we estimate that Buffett's leverage is about 1.6-to-1 on average. Buffett's returns appear to be neither luck nor magic, but, rather, reward for the use of leverage combined with a focus on cheap, safe, quality stocks. Decomposing Berkshires' portfolio into ownership in publicly traded stocks versus wholly-owned private companies, we find that the former performs the best, suggesting that Buffett's returns are more due to stock selection than to his effect on management. These results have broad implications for market efficiency and the implementability of academic factors.

* https://www.nber.org/papers/w19681

In other words, Buffett doesn't try to beat the market, he tries to be the market and waits for morons to walk in and lose to it. A bit like running a casino that lets customers set the odds but chooses which bets it accepts. That might be splitting hairs.

Reply on news.ycombinator.com