The Master of Robots Left AQR. Now He’s Coming for Wall Street

Discussion in 'Wall St. News' started by dealmaker, Oct 10, 2019.

  1. dealmaker

    dealmaker

    The Master of Robots Left AQR. Now He’s Coming for Wall Street (Bloomberg)
    The man dubbed the world’s top quant has a message for Wall Street firms lavishing millions on machine-learning programs: Most of you are doing it wrong. Robots could save active money managers from doom but the technology is struggling to live up to the disruptive promise, according to Marcos Lopez de Prado. The Spaniard spent years at the storied names of smart-money finance in his bid to shake up investing with self-learning robots. Now he’s travelling across quantland to deliver a reality check — selling his algorithms and expertise to all comers after an unexpected split from AQR Capital Management LLC this summer.
     
  2. any one heard successful story of self-taught ML in trading?
     
  3. shatteredx

    shatteredx

    Any evidence that de Prado has a positive PnL?
     
  4. Snuskpelle

    Snuskpelle

    His book Advances in Financial Machine Learning is mandatory reading for any quant.
     
  5. R1234

    R1234

    If he was that great. Why would he be peddling his stuff to institutions? Sounds like a hyped up academic relying on pedigree. Bet he's not even close to the same league as Jim Simons.
     
    shatteredx likes this.
  6. In what alternate universe is this?

    Among quants I know, few have read it. Among those who have, opinion is uniformly negative
     
  7. Snuskpelle

    Snuskpelle

    Obviously my opinion, hence alternate universe. No, the book is not perfect and there may well be better ones detailing overlapping issues, however I found it a good summary on a number of topics as with which he would be asking me "have you thought of this? this? this?" when trying to apply ML to quant problems. I would definitely have preferred to have read it five years ago before spending time on such myself (then and still a layman comparatively speaking, I should add).

    Mandatory might be too generous a term. Rhetorically I wanted to contrast with the trader replies above misunderstanding the nature of quant work, which although it may overlap with issues a trader may face most times doesn't (unless the trader is a self employed quant, that is).
     
  8. He didn't give clear examples of how quants are using machine learning the wrong way.
     

  9. What do you think ml in trading? You found the holy grail?
     
  10. ronblack

    ronblack

    You should read Kevin's book. It's the best. (Never engage in an argument with a know-it-all type)
     
    #10     Nov 6, 2019