BlackRock's Robot Stock-Pickers Post Record Losses
Read an interesting article regarding BlackRock robot stock-pickers posting record losses. As we all know, more and more Asset Management companies are turning to artificial intelligence to help with stock picking. In the case with BlackRock, Laurence Fink invested $78 billion in a quant team to distinguish the company from its peers.
Unfortunately for him, the article points out:
At least three of the quant strategies used by BlackRock’s Global Hedge fund platform have suffered losses greater than 10 percent in the year through November, according to the client update, a copy of which was seen by Bloomberg. That compares with an average return of 3.6 percent for quant funds, Hedge Fund Research Inc.’s directional quant index shows.
Does this mean the whole AI taking over the hedge fund/asset management industry is blown out of proportion? What does this mean for BlackRock? Would love to hear your thoughts.
Well, yes and no. The thing is, there are good AIs and there are terrible AIs. Why do people use Google over Bing/Yahoo/etc. Search, like quantitative investing, is a super complex problem and there are a lot of different ways of attacking it. Just because there are people putting money into shitty AIs doesn't mean that AI for investing doesn't have significant value. All it means is that BlackRock (among others) made a shitty AI. Of course, we need to look at returns year over year. Just because sometimes Bing will give you more relevant results than Google doesn't mean it does so consistently. The best quantitative funds do seem to beat the market more consistently than fundamental funds.
Source on the best quant funds beating the market more consistently than fundamental funds?
Renaissance Technologies
rentech, two sigma, DE Shaw, Citadel, millenium, etc.
the problem is many "fundamental" strategies have lagged because correlations were at an all time high and bottoms up work became less effective. combine that with people essentially mirroring indices and you had a recipe for disaster.
fundamental managers aren't innovating, and quant guys are. this is not a free lunch however, as the quant shops are consistently looking for an unexploited strategy, rather than relying on what worked in the 90s. they rely on an information advantage, a speed advantage, and hope that other arbitrageurs don't get the advantage before they do. traditional stock pickers wait for a good value, buy the stock, and hope their thesis works out.
the trouble is it's incredibly difficult to get access to the best quant funds (because they realize they have capacity constraints), so it's really a moot point.
The problem with AI is that you can't reasonably expect them to provide reliable returns until the platform matures.
These companies aren't walking down to some AI store to buy an off the shelf mature platform, the growing pains almost necessitate losses until they can collect enough data to make the platform stable.
This
BlackRock, being the world's leading ETF provider with iShares, will be fine. Their strategy is to diversify their huge leadership in passive management and smart beta towards active management, AI, robo-advisory, etc. Over time they will either win with their in-house solution or buy the independent winner.
I think it depends on what the AI is assigned to do. Are they investing or are they trading? If they are investing, what are they looking for and how are they set up to buy/sell? What's timelines are they utilizing? If they are trading, what market trends do they look out for? Are they engaging in high risk or low risk activities? Are they engaging in HFT on behalf of the fund? Are they using the AI to manipulate stock prices with aggressive buying/selling to drive the market? Does the specific AI allow for alerts and human input to help manage the fund? Who knows.
Without publicly available information to look at for insight, it's hard to make any judgment on the programs themselves and how they operate. They might not end up any better than your average manager's picks, sometimes you're lucky, sometimes you're not. You should also remember that the future has not played out yet and patterns that were previously viable for the AI to base judgment off of may not be suitable. Couple that with the volatility of a holding due to general sentiment/supply & demand and random unexpected macroeconomic events, I wouldn't expect them to be extremely consistent or perfect until there were a lot more variables within their control.
hhahahahaha
BlackRock's strategy is to spread a whole lot of bets then count on a few paying off. If they seeded 3 quant funds that failed, they probably have another 30 which "succeeded" or at least haven't failed yet. They do the same thing with fundamental active funds. BlackRock is the definition of closet indexer at least when it comes to their active equities business.
I could go on at length about statistical theory but.... Haha! Sucks to suck!
Looks like humans will be taking their jobs.
Trumps already helping good hard working Americans get their jobs back.
The Greatness is coming...
Just waiting for the "Money Monster" moment to happen
Voluptatem voluptas ad tempore eligendi. Id non aperiam corrupti quis. Est repellendus est magni dolorum nemo inventore. Non qui asperiores quia iusto. Est officia omnis quia enim aliquam velit tempore unde. Ut quasi voluptatem nihil et suscipit odio vel. Libero iure eum optio totam et.
Qui ratione nulla facilis in. Quibusdam ut corrupti eius et itaque. Itaque eligendi accusamus sed quia. Doloremque a mollitia ut ducimus culpa repudiandae qui.
Repellendus at velit qui ex ducimus reprehenderit. Quaerat rem voluptatem perferendis id dolorem facilis. In at consequatur est aut. Rerum quis earum quas consequatur sed similique. Quis consequatur odio excepturi harum pariatur voluptatem labore. Deserunt voluptas et ut placeat ut.
Corporis voluptatem qui error aliquam quidem minus nisi. Eum veniam facere voluptas ut possimus et qui.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...