[finstell]

> 20-30-40% per annum, for over 20 years

It is impossible, at least continuously. Nobody has ever managed to do that. You'd be lucky if you could beat the market by a few points on average over a 20 years time period. Compensation well for good performance does not make sense when you aren't penalized for losses.

[xiaoma]

That's just not true.

Joel Greenblatt's fund beat an annualized return of 40% from 1985 to 2006.

Carl Icahn got over a 30% per year annualized return from 1968 to 2011. That's almost 50 years!

[alasdair_]

Right, and a lottery winner who wins a $500MM jackpot on a $1 ticket has annualized return of even more than that, when annualized over the same timescale.

None of that matters if you can't pick the winning players in advance. With enough variance and enough players, someone will eventually have double-digit annualized returns over decades, it doesn't mean that they are necessarily superb investors.

[Gustomaximus]

The BIG difference is someone can go invest in Greenblatt or Icahn's fund with a reasonable expectation of these returns ongoing. Not so with a lottery winner.

[mehwoot]

Joel Greenblatt's new funds that he set up since returning in 2009 haven't done well, they have performed below the S&P 500 Index from 2009-2015:

http://www.forbes.com/sites/antoinegara/2015/07/31/value-gur...

[auntienomen]

What would you think if you saw someone win the lottery 2 months out of every 3 for 20 years?

[hodder]

Buffett has actually picked the winners in advance in his great essay:

http://gdsinvestments.com/wp-content/uploads/2015/07/The-Sup...

That said, a fund this size should be taking as passive of an approach as possible.

[greenleafjacob]

How would your position be falsified? You could always say that later someone will revert to the mean.

[Retric]

Simple, compare a large number of investors based on some criteria with the overall average.

AKA, if you think there are people that do better than average, then picking people who have beaten the odds for 10 years and see how they do over the next 10 years. Repeat over a few decades.

There are things that seem to work. The most common way to 'beat the market' is trading a low chance and ideally hidden chance of failure for inflated returns. EX: A 1 percent change of losing 95% of your investment should be worth lot's of money on good years. This is really appealing when investing other peoples money as you don't share in their downside.

[mehwoot]

Joel Greenblatt's new funds that he set up since returning in 2009 haven't done well, they have performed below the S&P 500 Index from 2009-2015:

http://www.forbes.com/sites/antoinegara/2015/07/31/value-gur...

[forgetsusername]

>Carl Icahn got over a 30% per year annualized return from 1968 to 2011. That's almost 50 years!

Citation? This would turn a $10,000 initial investment into $13BB. I have a hard time believing that.

[avn2109]

>> "...impossible..."

https://en.wikipedia.org/wiki/Renaissance_Technologies

"...famed for one of the best records in investing history, returning more than 35 percent annualized over a 20-year span..."

[tormeh]

People have done it before. It has always turned out to be luck. Fantastic track record until they cease being lucky.

So, cynicism and economic orthodoxy aside, that sounds like a really cool company. Has anyone tried just tossing a big dumb neural network on stock data and investigated whether it can make money? It sounds very obvious, but a quick googling returns little. But I guess the investment industry is pretty secretive by nature.

[hiddencost]

> Has anyone tried just tossing a big dumb neural network on stock data and investigated whether it can make money?

... yes.

[boulos]

One of our Solutions Architects wrote this up for amusement: https://cloudplatform.googleblog.com/2016/03/TensorFlow-mach...

[tomp]

You've no idea what you're talking about.

Renaissance technology is a quantitative trading company, meaning they use computers to trade; they probably do thousands of trades per day, and consistently make a profit. There's approximately 0% chance that it's just luck.

Now, you might say, they were lucky to stumble upon a strategy that works. You can also say that the strategy will stop working at some point (because of competition), so their "luck" will run out, and they might not get "lucky" in time to find a new strategy. But their past performance was most definitely not just "luck".

Btw, in most other professions (e.g. the arts, technical inventions, sports) we call this kind of "luck", "skill".

[auntienomen]

RenTec makes too many bets for their track record to be just luck.

[hueving]

Are you seriously asking if hedge funds that pay hundreds of thousands of dollars salary to top experts from all kinds of fields have investigated using machine learning?

[tormeh]

Obviously someone has, but I want to know the results.

[tomp]

You can't just "toss a NN to stock market data" and expect good results. There's too little historical data, you can't easily "generate" more data for the network to learn from, making it really easy to overfit. In other fields (e.g. computer vision) a lot of research has been focused on inventing techniques that prevent overfitting, thus enabling "learning" (i.e. generalization of patterns), such as dropout, convolutional neural networks, flipping/rotating images, etc. Very few of these techniques can be applied generally.

[eivarv]

Dropout can certainly be applied generally - it's useful as a regularization technique (especially in wide and deep networks) to combat overfitting in other fields than computer vision.

CNNs can be used in other fields as well.

[ktRolster]

It turns out to be rather difficult, because in most setups you see in the stock market, the market goes up roughly half the time, and down roughly half the time.

[groby_b]

https://xkcd.com/1570/

[robotresearcher]

There are millions of investors. The chances of a few outliers getting a long streak of returns are pretty good, even if they picked stocks at random. It's impossible to tell in advance who those lucky parties will be. But their existence is very likely.

[Retric]

Which is still less than someone who bought dell stock at IPO and held for 9 years.

[bkanber]

So?

Dell's market cap at IPO was $80M.

You can't just pick one good company to invest in. What would these funds do with the other, oh, $10Bn they need to invest?

[Retric]

These funds are not getting that kind of returns on 10B. Often you see "fund not seeking additional investment."

[bkanber]

That's my point. When you have so much money to invest you (are forced to) capture a more representative slice of the market and regress closer to the mean.

[melling]

And they used algorithms:

https://www.youtube.com/watch?v=QNznD9hMEh0

[adventured]

Continuously? Ok, perhaps that would be almost impossible, to pull off 20% for 20 years without missing a year.

However, Buffett and Soros managed to average above 20% annual returns over 30 plus years.

For example in the 1960s Berkshire returned 28.3% per year averaged. In the 1970s it returned 22.2% per year averaged. In the 1980s it averaged 39.1% (!) per year. In the 1990s it averaged 20.5% per year.

Nobody would hold their breath on another investor matching Buffett or Soros. It is in fact possible though. The big problem for the Norway fund is obviously the scale. Berkshire at $360b in market cap, will struggle perpetually going forward with keeping up with the S&P 500 over time (as is frequently noted by Buffett).

[webbore]

The thing about both Buffett and Soros is that they get deals that the general investor, or even really good and kinda famous investor, would never get. For example, Buffet did 300 million in unsecured loans (but with front-of-line payback) with Harley Davidson in 2009 at 15 percent, essentially to cover customer financing (i.e. cashflow) not because the company was in any real trouble. You and I would be lucky to find that sort of return from the riskiest loan, let alone a profitable manufacturing company. AND, Buffet could have made more than a Billion extra if he'd bought stock instead of debt, so even when he's wrong he's still getting a sweetheart deal.

[morgante]

It's important to note that this discussion is about a massive sovereign wealth fund, not the average investor.

I absolutely agree that the average person should stick to passive index funds. But if you have nearly a trillion dollars to invest, you'd be a fool to stick to passive strategies.

[executive]

"From 1994 through mid-2014 it averaged a 71.8% annual return."

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

[kristofferR]

Yeah, the oil fund averages just 0.25% above index, and that's without the extra cost of being actively managed factored in (although its huge size probably makes it impossible for it to be passively managed).

[010a]

Are you aware how much 0.25% of $800 billion dollars is? Now, go ahead and complain about a $60M compensation.

[robryan]

So at this stage factoring in the active management cost they are outperforming the market. Of course we don't know in the long run whether that will continue to be the case.

[ascorbic]

It's hard to beat the market when you're bigger than many markets.