Dell Algos

Hello everyone:

So after reading a recent post here on WSO:

http://www.zerohedge.com/news/2013-01-14/it-took-algos-4-seconds-fully-…-Dell-headline

I am confused about how these algos work.

I understand a program's ability to make trading decisions based on quantitative indicators in a market.
However, I am confused as to how an algo can make a decision based on soft info./ qualitative/ "words" data, and translate it into a quantitative trading decision.

To me, a program may be able to identify certain words or phrases, however the context and meaning of these words seems like something that only a human would be able to determine as credible or meaningful.

Is there anyone who can explain this to me?

8 Comments
 
trazer985you can assign values to words and read a data feed. plenty of services out there that offer the news feed which you can mine for data. It's expensive and pretty hard to get right. You need enough capital to do this across a lot of opportunities. 4 seconds is interesting as it is surprisingly slow for something that must be electronic.

Yeah, I guess I will just never understand. That kind of data just seems to subjective to me for a machine to be able to consistently and correctly analyze it. Of course bulletins with "going private" are pretty easy to get right, I'm just thinking about more complex situations that may never-the-less be just as or more important.

"History doesn't repeat itself, but it does rhyme."
 
trazer985you can assign values to words and read a data feed. plenty of services out there that offer the news feed which you can mine for data. It's expensive and pretty hard to get right. You need enough capital to do this across a lot of opportunities. 4 seconds is interesting as it is surprisingly slow for something that must be electronic.
That's because it is very likely someone had inside info and waited for the news to come out and bought it immediately afterwards. Electronic systems would be much faster
 

Nothing really comes to mind, and I know that most bulletins are quite straight forward. I'm just wondering how they assign values to these words given the vast depth of the english language and how different words in different contexts can mean different things.

"History doesn't repeat itself, but it does rhyme."
 

It's not as though algorithms are trading on every news story that hits the wire. As trazer suggests, there are likely only a few reports that trigger an algorithm, and very few of them take place during market hours.

"For all the tribulations in our lives, for all the troubles that remain in the world, the decline of violence is an accomplishment we can savor, and an impetus to cherish the forces of civilization and enlightenment that made it possible."
 

More importantly, I think you're overestimating how subjective the language in news articles is. If both of us can read the same news story and arrive at the same conclusion about what event is happening, then there is some objective meaning to the article. That's something that a computer can easily understand as well.

Ask yourself: how do you know that the article about Dell going private means that the shares will likely be purchased at a premium to the current market price? Once you answer that, you will know exactly how the computer arrived at the same conclusion.

"For all the tribulations in our lives, for all the troubles that remain in the world, the decline of violence is an accomplishment we can savor, and an impetus to cherish the forces of civilization and enlightenment that made it possible."
 
Best Response

Numquam aliquid alias ratione et et delectus. Quod quam laudantium sed culpa. Voluptatum deleniti numquam eos blanditiis assumenda eaque qui. Alias suscipit est voluptatem maiores aut.

Hic aliquam quibusdam dolorem ea. Veniam fuga saepe quia iure sit nihil. Suscipit aut alias ducimus iusto asperiores.

"History doesn't repeat itself, but it does rhyme."

Career Advancement Opportunities

August 2026 Hedge Fund

  • Point72 99.0%
  • D.E. Shaw 98.0%
  • Citadel Investment Group 97.0%
  • AQR Capital Management 96.0%
  • Magnetar Capital 95.0%

Overall Employee Satisfaction

August 2026 Hedge Fund

  • Magnetar Capital 99.0%
  • D.E. Shaw 98.0%
  • Blackstone Group 97.0%
  • Citadel Investment Group 96.0%
  • Two Sigma Investments 94.9%

Professional Growth Opportunities

August 2026 Hedge Fund

  • AQR Capital Management 99.0%
  • Point72 98.0%
  • D.E. Shaw 97.1%
  • Citadel Investment Group 96.1%
  • Magnetar Capital 95.1%

Total Avg Compensation

August 2026 Hedge Fund

  • Portfolio Manager (9) $1,648
  • Vice President (27) $464
  • Director/MD (11) $372
  • NA (9) $320
  • Engineer/Quant (83) $287
  • 3rd+ Year Associate (26) $284
  • Manager (4) $282
  • 2nd Year Associate (32) $253
  • 1st Year Associate (76) $193
  • Analysts (233) $178
  • Intern/Summer Associate (29) $145
  • Junior Trader (5) $102
  • Intern/Summer Analyst (276) $95
notes
16 IB Interviews Notes

“... there’s no excuse to not take advantage of the resources out there available to you. Best value for your $ are the...”

Leaderboard

1
redever's picture
redever
99.2
2
Secyh62's picture
Secyh62
99.0
3
BankonBanking's picture
BankonBanking
99.0
4
kanon's picture
kanon
99.0
5
DrApeman's picture
DrApeman
98.9
6
CompBanker's picture
CompBanker
98.9
7
dosk17's picture
dosk17
98.9
8
GameTheory's picture
GameTheory
98.9
9
Betsy Massar's picture
Betsy Massar
98.9
10
Mimbs's picture
Mimbs
98.8
success
From 10 rejections to 1 dream investment banking internship

“... I believe it was the single biggest reason why I ended up with an offer...”