Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Silicon Valley Hedge Fund Takes on Wall Street with AI Trader

bloomberg.com

Post reply on HN
Print view View original

91–100 of 111 posts

Re: Silicon Valley Hedge Fund Takes on Wall Street with AI Trader

#91

Post by kgwgk »

SEJeff wrote: post #88
kgwgk wrote: post #83

Earlier quoted context omitted.

> HFT or perhaps more accurately, Electronic Trading Electronic trading as opposed to what?

Click trading? Calling your broker and having them trade for you? Old school open outcry trading. Algorithmic Trading might be better?

Before you seemed to apply the term essentially to HFT only ("Most real electronic trading firms are putting those algorithms in hardware"), now it seems to include all the "systematic" trading. Or maybe only when fully automated? (I imagine low-frequency startegies might still require some human clicks).

And even if I call my broker, and he clicks somewhere to enter the order, it's still executed by some HFT agent somewhere. And the same applies to low frequency strategies, I guess (why bother with the execution details, just use a broker).

Re: Silicon Valley Hedge Fund Takes on Wall Street with AI Trader

#92

Post by SEJeff »

hendzen wrote: post #89
SEJeff wrote: post #87

Earlier quoted context omitted.

Yes, but people roll their eyes when you say that even if it is true.

Fair enough. If you're not in the nanos what is the point of even bothering with FPGAs and all the pain involved? CPUs can handle single digit micros just fine.

Agreed, if you're ever in Chicago look me up. Would be fun to grab beers with like minded techies (and that goes for anyone really).

Re: Silicon Valley Hedge Fund Takes on Wall Street with AI Trader

#93

Post by cityhall »

bfrog wrote: post #26

Huh, hasn't wallstreet being doing this, and even putting these things into FPGA's for nearly a decade now? Hell I wouldn't be surprised if they're printing ASICs with their algos.

Not all algorithmic trading is high-frequency where you need to shave microseconds to beat someone else's order. If you're predicting actual price movements instead of order book fluctuations there's no reason to use hardware.

Re: Silicon Valley Hedge Fund Takes on Wall Street with AI Trader

#94

Post by hendzen »

SEJeff wrote: post #92
hendzen wrote: post #89

Earlier quoted context omitted.

Fair enough. If you're not in the nanos what is the point of even bothering with FPGAs and all the pain involved? CPUs can handle single digit micros just fine.

Agreed, if you're ever in Chicago look me up. Would be fun to grab beers with like minded techies (and that goes for anyone really).

Definitely - not there very often but will keep in mind!

Re: Silicon Valley Hedge Fund Takes on Wall Street with AI Trader

#95

Post by BickNowstrom »

pdog wrote: post #9

> Thanks to increases in computing power, Sentient can squeeze 1,800 simulated trading days into a few minutes. How is it possible to generate seven years of convincing sample data from historical trading data without overfitting your models?

> How is it possible to generate seven years of convincing sample data

From the article:

> These genes are tested by giving them hypothetical sums of money to trade in simulated situations created from historical data.

So they simply use historical data (going 7 years back). This is fed into a simulator (a trading day where you have access to the data of all days leading up to the simulated trading day).

> without overfitting your models?

Traditionally backtesting is used: http://www.investopedia.com/terms/b/backtesting.asp

Care should be taken to run the tests over significant periods of time (to reflect changing market conditions) and to have a final out-of-time holdout set to lessen the effects of picking "winners" that were winning purely due to chance (introduced by cherry-picking winners from thousands of models).

Re: Silicon Valley Hedge Fund Takes on Wall Street with AI Trader

#96

Post by brilliantcode »

edblarney wrote: post #81
brilliantcode wrote: post #66

Earlier quoted context omitted.

This is just an anecdotal experience but according to a former hedge fund manager, insider trading is an open secret in the industry. You really can't expect to gamble away billions of investor's money without an edge. Information is the only possible edge that is closely guarded in the markets. I don't know how true that is, he's a washed up piece of shit that wall street chewed out but has the eyes and ears of gord…

Big funds will have direct access to CEO's and execs. They'll sit down for an interview. Technically speaking, everything that the CEO will say has to be public information. It has to be above bar. But sitting in the room, being right there, one might easily be able to glean more information than is actually public. Ergo - and edge. And it's not quite illegal. So that is a form of fairly above board 'soft inside' inf…

How many retail investors get face time with CEO of Kraft? It's a pretty rigged zero sum game for the rest with much of the edge in the hands of the few.

I actually started to treat trading very much like the casino. I just do it to make news reading interesting. Of course the mental pictures I form in my head are influenced by what I read. I use my gut....

Therefore retail trading is a social acceptable gambling.

brb shorting TSLA.

Re: Silicon Valley Hedge Fund Takes on Wall Street with AI Trader

#97

Post by CuriouslyC »

BickNowstrom wrote: post #69
CuriouslyC wrote: post #24

The problem with algorithmic trading is that markets are complex systems. As an algorithm is adopted, it changes the dynamics of the market, and the result the assumptions it was based on tend to no longer hold. Note that humans trade via implicit algorithms, which is part of the reason why consistently outperforming the market is highly unlikely. In my opinion, the whole idea of investing to try and maximize profits…

Let's assume your algorithm moves the market: 1) This is not a big problem: Lots of machine learning and control theory involves dealing with feedback loops. 2) If it moves the market in a predictable way, you can use this to make money. The Western world is capitalist, not a Platonic ideal. Even if you strongly feel that people should have different reasons to invest, you will not be able to sway them (unless you ca…

Point #2 is what I'm talking about. Any algorithm that is going to make money long term on a large scale has to be inherently unpredictable, which is difficult if it is an algorithm that predicts the future state of the market from past data.

Sure, the world is less than perfect in a lot of ways. In most cases we try and fix it. Why is capitalism the one way where we say "oh well, that's just how things are" ??

Research has demonstrated that barring insider information, random stock picking often outperforms "experts".

I believe the market is fairly efficient, and people who make money are either lucky or are leveraging short term information differentials, which in today's hyper-connected society are going to become increasingly rare (barring insider information, again).

Re: Silicon Valley Hedge Fund Takes on Wall Street with AI Trader

#98

Post by fokov »

mifeng wrote: post #37
65827 wrote: post #29

Earlier quoted context omitted.

I mean sure, but take glassdoor with a huge grain of salt, people make up crap to post on there all the time.

After a certain number of reviews, Glassdoor becomes a more honest reflection of a company's actual state than anything else. It's certainly more truthful that marketing fluff pieces like this.

I agree as well. I've noticed many problems in my current organization but it is really telling when people are complaining about the same things I'm seeing. The problems really exist, people have fought to fix them and failed, management either doesn't care or want to fix them.

Yes there will always be people that were fired writing things up as well as others just lying (competitors). Like you said, at some point when there is enough smoke you should expect a fire.

Re: Silicon Valley Hedge Fund Takes on Wall Street with AI Trader

#99

Post by pliny »

digler999 wrote: post #58
SEJeff wrote: post #21

The clickbait headline is amusing, but "Wall Street" has been doing this > 10 years. I'm genuinely not sure if they're trying to "beat wall street" or simply be yet another quantitative hedge fund that happens to use machine learning and neural networks to power their strategies. Most real electronic trading firms are putting those algorithms in hardware. Source: Have worked in HFT the past 10 years.

> Source: Have worked in HFT the past 10 years. OT question: how would I get into acquiring market data to try my own backtesting strategies ? Is there anyway to "get in" without putting up $5k + for the data ? I'm not green to this, I worked at a prop shop for a year writing market feed handlers. I know the dangers of overfitting data and how hard it is to make money (so I'm not deluded by a fantasy). I want to star…

TickData[1] are cheap and good for US data, I have had bad experience with them for EU data. They have a minimum of (IIRC) $1K but you can get a ton for that.

Some exchanges offer free historical data, for instance BM&F BOVESPA[2], trading in some markets (such as Brasil and China) is often difficult if you are a foreigner.

[1] https://www.tickdata.com/

[2] http://www.bmfbovespa.com.br/en_us/services/market-data/hist...

Re: Silicon Valley Hedge Fund Takes on Wall Street with AI Trader

#100

Post by digler999 »

jbmorgado wrote: post #90
SEJeff wrote: post #70

Earlier quoted context omitted.

Yes and I think you are willingly failing to understand reality from someone who has worked in technology in finance for a long time. HFT or perhaps more accurately, Electronic Trading, is more than simply passive trading and "doing something faster than everyone else". That is a part of it absolutely, but the industry also hires people with PHDs in quantum physics, or builds supercomputers at national labs to build…

Can you do HFT without a privileged connection to the trading system that beats most of the market players in access time? No. TL;DNR: You keep ignoring both the point of the comments and the point of the TFA. This is about market prediction, not technical exploitation of a privileged access to market.

> a privileged connection t

anyone can buy co-located rack space.

Post reply on HN