4 Crowdsourced Hedge Funds for Algorithmic Trading
A meritocracy is all about rewarding people on the basis of their abilities. This almost always turns out to be those people who work harder and take bigger risks. It’s what has made our present day society so great. Everything you see around you today is the result of a meritocracy that rewards those people who work the hardest and accomplish the most. The social media platforms people use to spew forth outrage about the “myth of meritocracy” and how “unfair the system is” fail to realize that it’s a meritocracy that has motivated the founders of said platforms to forgo evenings in front of the television in favor of 70-hour work weeks to build something that made them quite wealthy. The harder you work, the luckier you get. That’s one of the reasons why China is on track to eat America’s lunch. It is only by rewarding those people who work the hardest, that you can achieve great things as a society.
Perhaps no place is a meritocracy more evident than in the world of investing. An objective measure used to determine success in the investing world, is simply that when you initiate exposure to the market, you will exit that same exposure with more money than you initially started with. Of course we then need to take into account the risk that you took in order to do so, and compare that to the “risk-free” rate of return you could have received by letting your money “do nothing”. In order to compare investing strategies, we might use a set of criteria like the one seen below:
Let’s not let the verbiage get in the way of understanding some simple concepts. The simplest way to understand this is to think about “risk”. When we both make a “bet” on the market, we need to take into consideration the “risk” we each took. If you bought stock in McDonalds and I bought stock in Nvidia, and your position outperformed mine over the same duration, that would be a much more impressive feat than vice versa. It’s easier to make money with Nvidia and it’s also easier to lose money with Nvidia because the stock is more volatile than McDonalds. That volatility is also referred to as “risk”. You took less risk with Mickey D’s than I did with Nvidia and you had a better return. Of course we can take into account things like “risk adjusted returns” but you get the point we’re trying to make here. Risk is a key component when comparing investments strategies.
Now that we understand how you might compare different investment strategies, we’re in a much better place to talk about a crowdsourced investing platform that lets people compare investing strategies and be rewarded for building the “best” investment strategy in true meritocratic form. One company that’s best known for providing such a platform is Quantopian.
Founded in 2011, Boston, Massachusetts startup Quantopian has taken in $48.8 million in funding from big name investors like Bessemer Venture Partners and Andreessen Horowitz to build a platform that “inspires talented people everywhere to write investment algorithms“. All you need is a browser and you can code your own investment algorithm and then “backtest” it over 15 years of minute-by-minute data to see how it performs. Backtesting means that you can take a strategy and then see how it performs by running it over a period of historical stock market data. The very best algorithms are allocated real money to invest and the creators are given a share of the profits (around 10%).
To get started is quite easy. You can write your algorithm in a browser and then backtest it, for free, over 15 years of minute-level US equities data.
The best algorithms are then chosen to be applied against the market using real capital.
While the idea sounds solid on paper, it turns out that the devil is in the details. Late last year, the CEO of Quantopian departed citing “disappointing results” for the fund they launched in April 2017. According to an article by Fortune, “tens of millions of dollars” were “allocated to more than 15 algorithms created by members of Quantopian” with one algorithm given more than $3 million in funding.
Hedge fund manager Steve Cohen said back in 2016 that he would allocate up to $250 million to Quantopian algorithms. That’s a drop in the bucket compared to his total net worth which is said to be somewhere around $13 billion. With more than 160,000 users on the platform hailing from nearly every country in the world, it’s safe to say that he’ll at least be able to identify some talent that he can hire to help manage his hedge fund, Point72. It’s also safe to say that he’s probably able to fend off any competitors out there of which there are at least two; Quantiacs and QuantConnect.
Founded in 2014, Silicon Valley startup Quantiacs has taken in $2 million in funding to make “algorithmic trading accessible“. The business model is quite simple. The best three trading algorithms developed on the platform are allocated $1,000,000, $750,000, and $500,000. The creators of each algorithm get to pocket half of the performance fees. It’s similar to how hedge funds work in the real world. If the algorithms don’t make money, you don’t either. The last competition to wrap up was in August of 2017, where they chose three algorithms from a total pool of 492.
Founded in 2011, New York startup QuantConnect has taken in $1 million in funding to develop a system that “breaks down barriers to algorithmic trading by providing market data and a cluster computer so any engineer can quickly design an algorithmic trading system“. If you want to trade on live systems, there’s a subscription of $20 a month. You’ll be motivated to pay that because hedge funds might pay you a license subscription for your algorithms if they generate some promising alpha, regardless of the returns they generate.
The platform also lets you trade foreign currencies, futures, options, and even cryptocurrencies.
We’ve presented you with three platforms that let humans compete with one another to see who can build the best trading algorithms, but what about artificial intelligence (AI)? In past articles, we’ve looked at how you might use artificial intelligence for trading and how the best returns in the world come from a fund that uses artificial intelligence. Just how long will these platforms be in existence with AI dominating humans in just about everything? As it turns out, Quantopian already uses machine learning to improve trading strategies.
The simplest way to explain this is by reverting back to something in finance we call “factors”. A simple factor might be “size”. Stocks of a similar size tend to perform in a similar fashion. This is similar to how stocks of a similar industry perform in a similar fashion. If Intel has a bad earnings report and the stock falls unexpectedly, you can probably expect that the entire universe of chip making stocks will also have a bad day. More complex factors will take into account multiple data points. Consequently, complex investment strategies of the type that will outperform the market will use many different factors. The optimal weighting of these factors can be “optimized” by using machine learning. That’s an oversimplified explanation, but you can read this article from Quantopian if you’re interested in the nitty gritty details. This of course is a great segue into our last company, Numerai, which operates a bit differently from the rest.
Founded in 2015, San Francisco startup Numerai has taken in $7.5 million in funding to develop a platform that “regularizes financial data into machine learning problems for a global network of data scientists“. While the other platforms provide competitive environments to operate in, Numerai offers a collaborative platform. Numerai is unique in that they crowdsource problem solving which more represents “crowdworking“. The startup takes abstract financial data and then gives it to data scientists who do not know what the data represents. According to Richard Craib, Founder of Numerai:
Every data scientist on Numerai is solving the same problem using the same underlying features. But every data scientist approaches the problem in their own unique way. With many different solutions to the same problem, Numerai is able to combine each model into a meta model just like Random Forests combines decision trees into a forest.
Another unique feature of the platform is that the data scientists who solve these problems are compensated with the platform’s own cryptocurrency. Looking at the individuals that Numerai counts as investors and advisers is impressive:
It’s not hard to see how this platform is a great way to find the world’s best talent given how scarce data scientists are these days. While the world’s best hedge funds have maybe 200-300 data scientists, what about a hedge fund with 100,000 data scientists? That’s what Numerai is all about. More information on their business can be found in an article by the Founder titled Super Intelligence for The Stock Market.
For now, machine learning is being used to compliment strategies developed by humans so you still have a chance to take part in the meritocracy that these four crowdsourced hedge funds represent. If you think that the world of finance is somehow keeping your brilliant mind at bay, now is your chance to show everyone what a genius you are.
Pure-play disruptive tech stocks are not only hard to find, but investing in them is risky business. That's why we created “The Nanalyze Disruptive Tech Portfolio Report,” which lists 20 disruptive tech stocks we love so much we’ve invested in them ourselves. Find out which tech stocks we love, like, and avoid in this special report, now available for all Nanalyze Premium annual subscribers.