AI powered active management (AIEQ)

[NYEXPAT]

I might be the only one here into this stuff and at first I was a bit skeptical. I first jumped in at the beginning of the year and have to say the first six months as "Watson" was learning, were pretty disappointing. As we are coming up on the one year anniversary, I am very pleased with the performance and have doubled down today.

Anybody else own this?

 

[ERD50]

No, but it looks interesting. Looks like it was tracking the market for a while, and now is surpassing it. Is that because it is 'learning', or just going more aggressive in an up market, or random variation, or??

Can the AI 'know' what do do in different markets, and different companies coming and going? Maybe.

 

[Oz investor]

I might be the only one here into this stuff and at first I was a bit skeptical. I first jumped in at the beginning of the year and have to say the first six months as "Watson" was learning, were pretty disappointing. As we are coming up on the one year anniversary, I am very pleased with the performance and have doubled down today.

Anybody else own this?

i am still in the skeptical camp ( despite my geek hobbies )

i still consider myself a novice investor after 7 years and don't think "AI' will be a 7 times quicker learner than me in volatile markets ( a rising tide lifts all boats )

 

[Running_Man]

The fund is run by a young hedge fund potential manger who programmed his hedge fund ideas and theories into a computer program to read online news stories to determine which stocks are the best to buy.

WHY CONSIDER THE AI POWERED EQUITY ETF
• Amount of market data processed is unmatched - over a million
market signals, news articles, and 6,000 US companies analyzed daily
• Automated data driven investment process that removes significant
human bias and errors
• Active management that combines fundamental, technical, and
proprietary investment efficiency analysis to identify companies with
high opportunities for long-term growth
• Artificial intelligence and machine learning capabilities continually
build upon the financial knowledge base driving an investment system
that perpetually grows in value

Points one and three seem to be at odds with each other, reading millions of pieces daily holding about 50 stocks and trading to be in the best for the long term would seem to require constant trading of stocks based on long term outlook on a short term basis. Of course noone knows the program so it is unable to know what the basis is,

Each day, the EquBot Model ranks each company based on the probability of the company benefiting from current economic
conditions, trends, and world events and identifies approximately 30 to 70 companies with the greatest potential over the next twelve
months for appreciation and their corresponding weights, while maintaining volatility (i.e., the range in which the portfolio’s returns
vary) comparable to the broader U.S. equity market. The Fund may invest in the securities of companies of any market capitalization.
The EquBot model recommends a weight for each company based on its potential for appreciation and correlation to the other
companies in the Fund’s portfolio. The EquBot model limits the weight of any individual company to 10%. The Fund’s investment adviser utilizes the recommendations of the EquBot Model to decide which securities to purchase and sell,
while complying with the Investment Company Act of 1940 (the “1940 Act”) and its rules and regulations. The Fund’s investment
2
adviser anticipates primarily making purchase and sale decisions based on information from the EquBot Model. The Fund may
frequently and actively purchase and sell securities.

The fund does not make it easy to figure out the portfolio turnover but from the prospectus:

Portfolio Turnover Risk: The portfolio managers may actively and frequently trade securities or other instruments in the Fund’s
portfolio to carry out its investment strategies. A high portfolio turnover rate increases transaction costs, which may increase the
Fund’s expenses. Frequent and active trading may also cause adverse tax consequences for investors in the Fund due to an increase in
short-term capital gains

I do not know what happened in late April and Early May of this year but during that 6 week period the fund outperformed VTI by about 7 percent which accounts for it's gain over VTI which is presently about 1.75 percent after a little over a year giving some of that over performance back recently.

With a high expense ratio at .75% and likely high capital gains this is best probably for a tax sheltered account. But the investment is basically a very smart guy's computerized model of stock selections, without revealing his criteria. The idea that it gets "smarter" is most likely due to recalculation of weights of variables in relation to their correlation to market performance. Without knowing the strategy this is a pure bet on Art Amador's stock market modelling ability.

I find it interesting it "reads" all published financial stories and non-financial stories, which themselves are more and more written by computers, so it is basically computers determining what facts are worth publishing for other computers to determine impact of these "facts". I am more interested in the trade data of the fund than I am in the fund.

 

[OldShooter]

 

[Oz investor]

The fund is run by a young hedge fund potential manger who programmed his hedge fund ideas and theories into a computer program to read online news stories to determine which stocks are the best to buy.


Points one and three seem to be at odds with each other, reading millions of pieces daily holding about 50 stocks and trading to be in the best for the long term would seem to require constant trading of stocks based on long term outlook on a short term basis. Of course noone knows the program so it is unable to know what the basis is,



The fund does not make it easy to figure out the portfolio turnover but from the prospectus:


I do not know what happened in late April and Early May of this year but during that 6 week period the fund outperformed VTI by about 7 percent which accounts for it's gain over VTI which is presently about 1.75 percent after a little over a year giving some of that over performance back recently.

With a high expense ratio at .75% and likely high capital gains this is best probably for a tax sheltered account. But the investment is basically a very smart guy's computerized model of stock selections, without revealing his criteria. The idea that it gets "smarter" is most likely due to recalculation of weights of variables in relation to their correlation to market performance. Without knowing the strategy this is a pure bet on Art Amador's stock market modelling ability.

I find it interesting it "reads" all published financial stories and non-financial stories, which themselves are more and more written by computers, so it is basically computers determining what facts are worth publishing for other computers to determine impact of these "facts". I am more interested in the trade data of the fund than I am in the fund.

+ 1

please remember computers ( both hardware and software ) make errors as well ( and a human created them both either perfectly or imperfectly )

a plus for the human is they can realize mistakes made quickly and correct the damage quickly ( maybe even fix the problem in the medium term )

 

[ERD50]

...

a plus for the human is they can realize mistakes made quickly and correct the damage quickly ( maybe even fix the problem in the medium term )

Yes, I could imagine some feedback loop that ends up feeding on itself and running the account to zero!

OK, probably not that bad, but I think I'll just watch.


-ERD50

 

[Oz investor]

i should be fair and admit that in some places the human is not so responsive either ( regardless of how much they are paid )

noise ( media or digital ) is still a problem otherwise the guy with the programmable mouse would still be happily trading

most of the time i will be backing the human ( especially myself ) for complex choices the human will be punished for bad results

so far a 'smart beta ETF that uses equal weighting ( and 2 monthly rebalancing ) is about as AI active as i find attractive ( but i haven't bought in yet )

 

[TwoByFour]

A few reasons why I am skeptical....

A. Wall Street has been hiring legions of mathematician and programmers for years to create algorithms that will "beat" the market. So far they have been unsuccessful. How is this different from that?

B. AI is the latest hot technology. Putting "AI" into the name, or in the marketing literature is sure to gain some buzz. This smells like a bit of marketing hype to me.

C. If somebody had actually created a sure fire way to beat the market with AI, they could make far more money by patenting it and licensing it. And if they don't patent it, everyone else will soon be doing it and it will no longer be an advantage. The advantage will effectively be arbitraged away.

D. Read A Random Walk Down Wall Street on how hard it is to beat the market.

 

[njhowie]

A. Wall Street has been hiring legions of mathematician and programmers for years to create algorithms that will "beat" the market. So far they have been unsuccessful. How is this different from that?

C. If somebody had actually created a sure fire way to beat the market with AI, they could make far more money by patenting it and licensing it. And if they don't patent it, everyone else will soon be doing it and it will no longer be an advantage. The advantage will effectively be arbitraged away.

Don't tell that to Jim Simons.

https://en.wikipedia.org/wiki/James_Harris_Simons

 

[TwoByFour]

Don't tell that to Jim Simons.

https://en.wikipedia.org/wiki/James_Harris_Simons

Yeah Simons is a rich hedge fund guy, and smart. But his RIEF fund has not beat the market. Over the last 10 years it has grown 6.2% but the total US market has grown 12.1%.

 

[Running_Man]

Yeah Simons is a rich hedge fund guy, and smart. But his RIEF fund has not beat the market. Over the last 10 years it has grown 6.2% but the total US market has grown 12.1%.

This is a fund run not by Simons but people who came on after he retired. Simons has averaged 71.4 percent before fees per year from 1994 to 2014 a twenty year period in his Medallion fund, which is closed now to only employees and original investors. It has averaged 37 percent per year since it's inception. In 2008 Medallion made over 80 percent (REIF fund lost money). From 2001 to 2014 the S&P500 has beaten the Medallion fund exactly once. The RIEF fund did not even exist until 2005.

 

[OldShooter]

Don't tell that to Jim Simons.

https://en.wikipedia.org/wiki/James_Harris_Simons

The main theme of Nassim Taleb's Fooled By Randomness is where a person gets lucky and, from that, concludes that he is a genius. It is very similar to the phenomenon we see every day in the financial press, where the current most-lucky forecaster is anointed as the current genius.

I never heard of this guy, but from the Wikipedia bio it looks like that he is a very smart guy in mathematics and one who has been somewhat lucky in benefiting from the random process that is the market. A perfect example is his two funds they talk about, Renaissance and Medallion, with wildly different performance. Assuming both are run honestly, that kind of thing is exactly what you would expect from sampling a random process. One fund got lucky, one didn't.

 

[njhowie]

OldShooter, think what you like, I'm not looking to convince you or anyone else of anything. I've simply provided a counter to TwoByFour's post. That you've never heard of him or what he has accomplished means little - type his name in to google and you get 12 million results. Anyone who has worked on Wall Street on the front lines knows who Simons is and about Renaissance.

He stays out of the public eye, and that's likely contributed to his success to a certain degree.

If you have some time, go to Youtube, type his name in, and watch some of the videos. You may have a change in your thinking.

 

[OldShooter]

OldShooter, think what you like, I'm not looking to convince you or anyone else of anything. I've simply provided a counter to TwoByFour's post. That you've never heard of him or what he has accomplished means little - type his name in to google and you get 12 million results. Anyone who has worked on Wall Street on the front lines knows who Simons is and about Renaissance.

He stays out of the public eye, and that's likely contributed to his success to a certain degree.

If you have some time, go to Youtube, type his name in, and watch some of the videos. You may have a change in your thinking.

I will look. I was typing when your post came it so I missed it.

 

[Oz investor]

Is This the Silver Bullet That Will Save Active Management?

https://www.institutionalinvestor.c...ilver-Bullet-That-Will-Save-Active-Management

i still remain skeptical on this

i use a mix of relatively passive fund-manager styles and active styles ( buying more when that style is out of favour , but not a complete mess is still has to be good manager , not an error prone one )

 

[MRG]

Yes, I could imagine some feedback loop that ends up feeding on itself and running the account to zero!

OK, probably not that bad, but I think I'll just watch.


-ERD50

Don't be sure.

I listened to a heartwrenching tale told by a father/son team. Dad's money and jr, a laid off programmer, hatched a scheme to trade VIX futures, programmatically. There were bugs and then the 2008 stuff happened. Different times and technology, SSDD.

It was at an investment discussion but it felt like a 12 step meeting.

 

[Oz investor]

personally i would like to retain some control over investment decisions

how would that combine with the AI program ??

as a home project ( AI designed for me exclusively ) i foresee my unreliable ( speed wise and connectivity ) ISP as a deal-breaking problem ( potential for software/hardware errors would also make it unattractive to me )

would i use AI as a filter for my news-feeds ( NOT ticker-codes watching ) , now THAT might be useful

 

[OldShooter]

OldShooter, think what you like, I'm not looking to convince you or anyone else of anything. I've simply provided a counter to TwoByFour's post. That you've never heard of him or what he has accomplished means little - type his name in to google and you get 12 million results. Anyone who has worked on Wall Street on the front lines knows who Simons is and about Renaissance.

He stays out of the public eye, and that's likely contributed to his success to a certain degree.

If you have some time, go to Youtube, type his name in, and watch some of the videos. You may have a change in your thinking.

Ok, I looked around a bit. It's a puzzle, frankly. I could not find any hard information on the high-performing, employees-only, Medallion except, apparently, what the fund discloses voluntarily. The vast difference between it and Renaissance is also unexplained. But the guy certainly seems to have made a lot of money.

I'm not too interested in pursuing it further. Maybe he's really a black swan. But the problem with winners like this is that you can't pick them ahead of time, only in the rear view mirror. As the buyers of the Renaissance fund apparently found out.

Interesting, though.

 

[NW-Bound]

I just recalled about this AI-powered ETF, and checked into it to see how it has been doing.

Since the inception in Oct 2017, the fund has returned 2.65% vs the S&P 13.5%. And the comparison will look worse for the fund if we consider its dividend yield of 0.5% vs. 1.8% for the S&P.

What makes it even worse is that at the bottom of the market rout late last year, the fund was down nearly 15% from its inception, while the S&P was down less than 7% in the same period.

Using AI to trade stocks is not so easy, it appears.

 

[ERD50]

I just recalled about this AI-powered ETF, and checked into it to see how it has been doing.

Since the inception in Oct 2017, the fund has returned 2.65% vs the S&P 13.5%. And the comparison will look worse for the fund if we consider its dividend yield of 0.5% vs. 1.8% for the S&P.

What makes it even worse is that at the bottom of the market rout late last year, the fund was down nearly 15% from its inception, while the S&P was down less than 7% in the same period.

Using AI to trade stocks is not so easy, it appears.

Thanks for bringing this back up.

Here's a better tool that includes divs, set the scrub bar to "ALL".

https://stockcharts.com/freecharts/perf.php?VTI,AIEQ

A little different numbers, cannot all be attributed to divs I don;t think, so might need to dig a bit to confirm, but the same story.

I might be the only one here into this stuff and at first I was a bit skeptical. I first jumped in at the beginning of the year and have to say the first six months as "Watson" was learning, were pretty disappointing. As we are coming up on the one year anniversary, I am very pleased with the performance and have doubled down today.

Anybody else own this?

"today" being 09-28-2018.

Take the above link, enter 225 (trading days) into the scrub bar.

AIEQ down 3.53%,

VTI up 1.21%

AIEQ has lagged VTI the entire times since 9-28-2018.

Still seems that applying no intelligence is the way to go, just B&H.

-ERD50

 

[NW-Bound]

Thanks for bringing this back up.

Here's a better tool that includes divs, set the scrub bar to "ALL".

https://stockcharts.com/freecharts/perf.php?VTI,AIEQ

A little different numbers, cannot all be attributed to divs I don;t think, so might need to dig a bit to confirm, but the same story.

Thanks for double-checking my numbers. They are way-off due to my negligence.

AIEQ did not trail the S&P as badly as I reported earlier. There was a huge cap gain that was distributed late last year of around 10%. AIEQ still trails the S&P, but not by more than 10% as I mentioned earlier.

 

[OldShooter]

It takes 5-10 years of data before performance can even begin to be evaluated, but yes it is common to confuse stock price history with total return. The press does it all the time with market averages like the Dow and the S&P. This significantly understates actual performance.

 

[Lsbcal]

I might be the only one here into this stuff and at first I was a bit skeptical. I first jumped in at the beginning of the year and have to say the first six months as "Watson" was learning, were pretty disappointing. As we are coming up on the one year anniversary, I am very pleased with the performance and have doubled down today.

Anybody else own this?

Is there any reason to think the algorithms used in this software approach are superior to other software approaches or other active management techniques? Suppose the term "AI" were replaced with Fortran. Would it be less sexy then?

Isn't "AI" a catch all for a broad range of software techniques? Yes, I'm suspicious of the use of sexy tech terms.

Fundamentally I would never invest in a black box approach. One would be elated if it outperformed the market and personally I'd be jealous of it's success. But as soon as it starts to falter one would be dealing with buyer's regret (at least this one would).

 

[NW-Bound]

It's been 4 years since this thread was started. Because of the thread on SDC (Self-driving car) which uses AI (Artificial Intelligence), I recall this AI-driven ETF and look up its performance.

If you invested $10K in AIEQ in Jan 2018, its inception almost 5 years ago, you now have $13,645 vs. $16,538 if you chose VFINX (Vanguard S&P 500). Both numbers include dividend reinvestment.

In terms of CAGR (Compound Annual Growth Rate), AIEQ is 6.53% against 10.77% for S&P.

Is the AI stock trading computer getting better? It does not appear so, as the 2022 YTD performance is -25.32%, against the drop of -13.22% by the S&P.

Source: https://www.portfoliovisualizer.com