Comment by pmart123 | Hacker News Reader

Comment by pmart123 | Hacker News Reader

Parent Full thread

Personally, I struggle to see the competitive advantage Quantopian brings. They use retail brokerage platforms to facilitate trading, which rules out anything close to HFT. Then, they are tied to any financial data vendor (Morningstar in this case) to not offer too much visibility on the underlying data. As others have mentioned, this makes it tough to validate aspects like adjusted vs. as reported earnings, how delistings are handled, etc. From my experience, getting/making sure the data is accurate is a ton of work, even if it is from good sources and you can see all the actual data. The moment an investor/trader on the platform gets traction is the moment they want Compustat data, exchange data, Bloomberg for fixed income, and will trade through Instinet/Flextrade, etc. The moment the platform is successful is the moment Morningstar could pull the rug out from them. If someone has better understanding or knowledge on Quantopian in particular, I'd be interested to hear why.

My understanding is that Quantopian has no interest in being a platform for HFT. It's there to democratize trading strategies and test them in a sound way. They do a lot to have a good backtesting platform and clean the data that is available in the platform. If I wanted to automate a strategy, I'd have to figure out the Robinhood API and basically recreate what they have. Instead, I can use their platform, their backtests, their free data, and build strategies that efficiently lose money.

I think the other goal being that if you create a profitable strategy, you can enter it into competitions, trade with other people's money, and make a profit. I think the platform is a value add, I have an account and use the research notebooks. Some day when I more spare time I'll make some trading algorithms.

It sounds like day trading for script kiddies. There's no way to develop or execute a sound strategy with that level of access.

It's like a slot machine tournament. No one has an edge but some percentage of people will do well enough over a finite period to convince them to put real money behind their lucky algorithm.

I understand they don't, but the execution options could matter. Retail brokerage platforms don't actually send your orders to the exchange: http://www.schwab.com/public/schwab/nn/legal_compliance/impo.... If your turnover is very low, and your trade size is small, this shouldn't make an impact in liquid products, but only having market and limit orders can be very limiting. Granted, IB is probably the best retail platform, but it's not perfect. Even a smart beta strategy would likely have 4X turnover to an index. I do hope they succeed, and perhaps they begin offering a more open, premium offering to compete with Capital IQ, etc. I just have a hard time seeing how operating as a fund-of-funds essentially is a good business model considering the CAPEX they have to build the software. I could see it as a way to build, test, and then offer a premium back-testing platform or portfolio analysis/attribution tool once they've ironed out the bugs. There's a lot more money in that business than in fund-of-funds.

The main problem with Quantopian is that the data missing especially for delisted stocks, here are some examples you can test yourself: AA, CWH, NIHD, PANL, HTZ, DSL - I have a whole list I found very fast, just imagine what other mistakes could be there... And the other problem is that they are my competitor, so I would never give them my trading strategy or ideas I am testing at the moment...

The trading strategy stuff has been brought up before. They say when evaluating algorithms they only look at the alpha, beta, etc. and not the algorithm itself. While the system can see what you're running, I think they work pretty hard to let your code be yours. I'm definitely not very advanced in algorithmic trading, so I see how there could be some data issues that I wouldn't come across.

This is false. Quantopian sources their historical and real time data from Nanex, and the data is explicitly free of survivorship bias (i.e. they maintain data for delisted equities).

I don't know what you verified, but it directly contradicts my own experience and the FAQ: https://www.quantopian.com/faq#data-sources

To your second point, as another commenter said Quantopian doesn't see your algorithm unless you give them permission, as outlined in the terms of service (which are very readable, by the way).

NB: I am not unaffiliated with Quantopian and don't use the platform for my own trading, but I have tried it out before.

Please go back and check and you'll se the data is missing for the symbols stated above, and I am sur for many more based on my limited test... everyone can see I am right by trying to request the data, nothing false about that...

> everyone can see I am right by trying to request the data, nothing false about that...

Oh, they can? Well that's funny, because I just went onto Quantopian and checked for myself.

http://imgur.com/a/VwMUJ

http://imgur.com/a/VV68C

Gee, would you look at that...they are there.

This took me all of five minutes. Do you have anything else I can easily disprove while I'm at it?

I like your script naming scheme...

The problem is the data is missing for specific dates... I just can imagine what else is missing or it is wrong...

This is really trying my patience. You made a clearly incorrect claim without mentioning specific dates and challenged me to disprove it. When I did disprove the claim, you shifted the goalposts with this backpedaling remark about how only specific dates are missing. You continue to make these claims and put the burden of proof on everyone else because you can't be bothered to defend them when you're challenged.

I'm not searching for a needle in a haystack for you. Make a falsifiable claim, and make the entire claim, with the specific dates you're talking about, or stop this astroturfing against Quantopian and QuantConnect that you're doing.

My goal is to show people how start to think for themselves and hopefully create lifestyle business around trading or invest for themselves instead of using "buy and hope strategy" and how to do it without losing a single penny while learning and to show them the barrier for entry is pretty low... But you are right, why bother when every time I gave an idea, the day traders with the thousands of dollars data feeds are all over my posts... no good deed goes unpunished here at HN I guess :-(

Shitting on other people products with vague and inaccurate criticisms isn't a good deed, it's just bad form. People don't need you to "show them how to think for themselves." The other guy is right, either make a sound and specific criticism or don't say anything. You did exactly what he said, you made a bad argument which he refuted, then you moved the goal post so as not to appear wrong, while still leaving out the necessary detail to disprove your new vague claim, rather than admitting your mistake and making a better argument that included specifics about what is missing.

What dates do you see missing data for?

They're searching for "alpha" strategies. Then they buy/share profit with the creator... I have a hunch that they are feeding these good strategies to some type of ML in order to become more efficient and profitable.

Reply on news.ycombinator.com