How Ray Dalio built the world’s richest and strangest hedge fund (2011) | Hacker News Reader

How Ray Dalio built the world’s richest and strangest hedge fund (2011)

newyorker.com

38 pointsjimsojim22 comments

It seems to me that the continued debate over whether or not the market is random (or whether it is not), is very similar to the debate of nature vs nurture.

My instincts, and my conclusion is that there is an element of both randomness and non-randomness in the market. I would expect then that there would be both successful and non-successful actors who succeed/fail due to random and non-random factors. My gut further leads me to believe that having insight or knowledge of the market coupled with a very large bankroll would allow you to ride the random events/waves with prudent mitigation strategies. A hedge, if you would. My expectation also would be then you could measure almost any fund on some arbitrary timeline and argue that the traders either knew or didn't know what they were doing, and that because of random event XYZ they either failed to correctly predict market movement ABC over period IJK. Let's not even get into the MNO or DEF parts!

Whatever "culture" of radical honesty they have does not translate at all to other workplaces. My first experience with an ex-BW manager left me crying at the coffee machine all afternoon. just my n=1 anecdote.

I think they understand this. See the first minute of this video https://www.youtube.com/watch?v=ABB1pfi3ZpE

Ray Dalio has a 30 minute YouTube video:

https://www.youtube.com/watch?v=PHe0bXAIuk0

Great place to work.

All investment funds are a charade. The probability that a firm like Bridgewater exists with higher than average returns is not zero. Nothing about Bridgewater would indicate that they somehow 'get it' whilst the other funds don't. It's strictly probability, or more accurately, it's strictly a bell-curve.

Your conclusion is based on the assumption that the market is completely random and unpredictable. Which is demonstrably not true - high frequency trading, insider trading, general predictable events that affect the market.

The market is random and unpredictable when it's random and unpredictable. And it's the opposite when it's the opposite. There will always be a small segment of traders who will do well. Doesn't mean that they are the Oracle of Delphi. It just means that, in a large enough sampling, the probability of those people existing is not zero. There is nothing in HFT that supports an argument of predictability unless you are another HFT possibly. Insider trading is a crime specifically because it removes the randomness of the market so I'm not sure why that's listed. And if there were, as you say, general predictable events then that does nothing to support the idea that Bridgewater has any special access to those.

Dr James Simons says the market isn't random.

https://en.m.wikipedia.org/wiki/James_Harris_Simons

https://www.ted.com/talks/jim_simons_a_rare_interview_with_t...

Now I'm not sure who to believe, some random guy on the Internet, or a mathematcian who has consistently beat the market.

And yet Mandelbrot thought the market was random.

Now I'm not sure who to believe, one of the most important mathematicians in the last 50 years, or a dude who left the academic community to start a hedge fund.

Mandelbrot did not think the markets were random. He explicitly rejected the Efficient Market Hypothesis.

http://www.amazon.com/Mis-behavior-Markets-Benoit-Mandelbrot...

Maybe I'm missing something, but how are randomness and the emh related?

AFAIK random walk theory assumes the market is unpredictable and is consistent with the emh while other theories like the adaptive market hypothesis assume the opposite and are still consistent with emh.

If the EMH were true, price curves would always display maximum entropy, i.e. randomness, because there would be no spare redundant information that could be used to make predictions about the future. (This is based on Shannon's Communications Theory, but the maximal entropy bound applies to any system that mixes a predictable signal with random noise.)

It doesn't matter if you use an evolutionary explanation for price curves, as in AMH, or claim they're controlled by planetary alignments - because the prediction that price curves show maximum entropy is falsifiable regardless of possible causes.

And when it's tested, it is indeed falsified. See e.g.

http://www.turingfinance.com/hacking-the-random-walk-hypothe...

tl;dr There are standard tools for estimating entropy, and they all agree that markets aren't truly random. Therefore they can't be maximally efficient.

Quants make a living by mining the signal from the randomness. There's a lot of debate about the best way to do this, but there's no serious disagreement among quants that it's possible - and the people who make money by employing them tend to agree.

Interesting, I'm familiar with information theory but not in the context of financial markets. I'm a little confused though because I assume you're referring to the channel coding theorem (mixing a signal with random noise), but that seems like begging the question to me. Don't you have to make the assumption that the market is a predictable signal with noise for it to be applicable?

I'm sure people employed to predict the stock market believe they are able to predict the stock market, but as far as I'm aware it's an open question in academia.

Also, I should point out, although I do believe markets are random, the intent my original post was more to point out the parent's argument by authority.

IMO, the market is only non-random due to trader herd-mentality. Everyone using the same software, same algorithms, trend-analysis, etc. If everyone's software says that this particular point is a resistance/support level, then the market will potentially move because of this. But this strategy of waiting-for-the-dumb-traders-to-make-their-move has been practised for decades. But a single random event will invalidate all this herd-mentality in a second. And plus, from your Wiki link, it states: "[Simon's] models are based on analyzing as much data as can be gathered, then looking for non-random movements to make predictions". I don't see that as being earth-shattering considering that random events occur constantly. And how do they know the movements are non-random?

I'm not sure why you got downvoted. Survivorship bias is a well-known phenomenon. I would be interested in distinguishing between hedge funds that do well because some "necessarily must" and those that actually have some kind of advantage with their algorithms. But how can you actually tell which is the case?

I don't think you can. If on every trade you flip a coin, but intelligently limit your losses and ride your wins, you will beat every fund-manager on the planet. Nothing to do with smarts or algorithms. And even if you have good risk management, if you trade long enough, there will always be the perfect-storm 'meltdown' scenario. However, as I originally said, it's all probability and there will be a small group of managers who will never experience the meltdown (or the meltdown is waiting in their future).

> If on every trade you flip a coin, but intelligently limit your losses and ride your wins

What does this mean? Sounds like gambler's fallacy combined with a Martingale system

Flip a coin... heads you go long, tails you go short. Limit your losses and ride the wins. You'll kill the market (until a perfect-storm meltdown event, of course). Nothing to do with Martingale.

Could you elaborate on "Limit your losses and ride the wins"? Do you assume that market is trending upwards (until meltdown)? Do you suppose to make money this way in flat market?

I recommend the book "More Money Than God" which studies the history of hedge funds and in the end suggests that successful hedge funds are not simply statistical survivors.

I would agree if you said that “most investment funds are a charade” or that most investment funds aren’t worth the fees they charge. Though I’d ask for a citation that Bridgewater (and other positive outliers) fall within statistical expectations, because I believe the data would show that there are many more positive outliers than you would get from pure chance [1].

Also, how much diligence have you done to conclude “Nothing about Bridgewater would indicate that they somehow 'get it' whilst the other funds don't”? Have you ever evaluated their knowledge, resources, and culture vs. the market? Do you not believe they have the culture described, or do you not believe it will result in better investment returns than the market? I imagine that you don’t read Bridgewater’s Daily Observations [2]. If you did, I think you would be quite impressed and would attribute at least some of their returns to skill.

[1] Here are two references to a similar, but different, rebuttal: https://en.wikipedia.org/wiki/The_Superinvestors_of_Graham-a... http://www8.gsb.columbia.edu/alumni/news/superinvestors

[2] https://en.wikipedia.org/wiki/Bridgewater_Associates#Daily_O... Here is an example of some of their economic publishing http://www.bwater.com/Uploads/FileManager/research/how-the-e...

Reply on news.ycombinator.com