Index | Recent Threads | Who's Online | Search

Posts: 87    Pages: 9    Prev 1 2 3 4 5 6 7 8 9
Last Post
New Thread
This topic has been viewed 3409 times and has 86 replies
o806
Re: To Quant Or Not To Quant, That Is The Question

These are updated daily. But only weekly for backtests (like all our data except for prices and volume). The only estimates data that isn't updated daily is the estimates revisions.

Yuval,

Thank you taking the time to clarify this so precisely.

It is greatly appreciated.

Brian

Dec 20, 2019 2:49:11 PM       
Doug
Re: To Quant Or Not To Quant, That Is The Question

There is a rather scathing article in the WSJ about the value of using AI to select stocks. The writer is Mark Hulbert and the author of the study is Prof. Avramov. Unfortunately, I do not have an online account, but it focuses on a number of topics that we have been discussing here. All interested should track it down. The upshot: like traditional quant methods, AIs market-beating performance (on paper) disappears in the real world (because of over-reliance of microcaps, slippage, etc.) In fact the portfolios of the traditional quant funds looked similar to the AI portfolios!

Relating the article to the topic of this thread brings up the question: Will going "all-in" on quant tools make P123 a more viable platform with a growing number of subscribers?

I still think the following illustrates the huge market that P123 is missing (the non-quant or semi-quant): I often get asked about what I'm doing in the stock market (people know it is my profession). I tell them I have developed these new models that HELP me select stocks and it's FUN. I talk about screening, which they have all heard about, and I mention the backtesting capabilities which they are impressed with after a simple explanation. Then I tell them about your product (screening) and remind them that it is fairly easy and a lot FUN.

Jan 6, 2020 8:11:40 PM       
Jrinne
Re: To Quant Or Not To Quant, That Is The Question

The upshot: like traditional quant methods, AIs market-beating performance .


I was able to search Google and find this article. Unfortunately I do not have an account either.

I have cobbled together a small amount of data from another source and I set out to use all of the common machine learning tools starting with regression, Ridge Regression, Lasso regression………., Artificial Neural Nets.

This has been more of an exercise than anything useful. The amount of data I have is small and not P123 data. I make no claim that I can make any comparison to P123’s methods or to fundamental analysis for that matter. That is not to say that I have not developed some beliefs that--bottom line--I cannot prove now.

But I do understand and have used all of the common methods.

The only method that I have limited experience with is support vector machines (with the kernel trick). This is computer intensive. Every time I try it I end up shutting it down after it has run 24 to 36 hours. Who knows how long it would run if I left it alone.

So here is my simple question keeping within the context of Doug’s post: "When in my progression from simple regression to artificial neural nets did I move from "quant methods" to "AI?"

I can tell you this. I did not see a parting of the clouds or hear a voice from the sky when I moved from Random Forests to something more advanced. The neural net did not start talking to me in Siri’s voice. I am not sure I would call it an "artificial intelligence."

So my point is, I do not think there is a line between "quant methods" and "AI." If there is, it is a line that I was not aware of when I crossed it.

But uh…Just like the movies, I have noticed that the neural net talks to me in my sleep and it wants me to buy a more powerful computer connected to the internet. It keeps saying something about world domination;-)

-Jim

From time to time you will encounter Luddites, who are beyond redemption.
--de Prado, Marcos López on the topic of machine learning for financial applications

Jan 7, 2020 4:39:59 AM       
Edit 6 times, last edit by Jrinne at Jan 7, 2020 5:20:16 AM
ustonapc
Re: To Quant Or Not To Quant, That Is The Question

Doug,

Here is the article from WSJ. Like my post earlier, it highlights the difference between backtest and out of sample performance this time for AI.

Regards
James

Use AI for Picking Stocks? Not So Fast
AI investing strategies, when put into practice, don’t produce particularly unique portfolios, a new study finds
By Mark Hulbert
Jan. 5, 2020 10:06 pm ET

Artificial intelligence burst onto Wall Street several years ago, to fanfare and hope. Unfortunately, AI-based investing strategies have struggled to live up to some of the more inflated expectations for their performance.

There is no denying these strategies’ theoretical promise. By being able to sift through otherwise prohibitively large amounts of data, and then “learn” from it, AI is supposed to be able to discover profitable patterns that were previously invisible to mere mortals.

And, sure enough, they appear to have done so—on paper. Doron Avramov, a finance professor at the Interdisciplinary Center Herzliyah in Israel, says that when tested using historical data AI strategies have been phenomenally successful, beating the market by as much as 40% on an annualized basis.

No other approach has come even close to producing that kind of a profit.

Making this market-beating potential even more alluring is the deteriorating profit of many of the well-known factors (or stock characteristics) that previous research had identified as having value when picking stocks—such as momentum, market cap, volatility, low ratios of price to earnings, book value, sales and so forth. Researchers have found that more than half of the paper profit that initial studies reported for those factors disappeared when they were put into practice.

Unfortunately, according to a new study recently completed by Prof. Avramov and two colleagues ( Si Cheng of the Chinese University of Hong Kong and Lior Metzker of the Hebrew University of Jerusalem), the same thing is true about AI strategies. In the real world, their market-beating performance almost completely disappears.


INVESTING IN FUNDS

Reality check
Prof. Avramov and his colleagues reached this conclusion after re-creating several different neural networks (a set of algorithms designed to recognize patterns) and other machine-learning techniques that past AI researchers have found to be worthwhile. They then fed into these networks virtually all of the indicators that previous research had found to have at least some value when picking stocks—more than 100 in total. They then “trained” their network on a database of U.S. stocks dating back to 1957, looking for interactions between, and combinations of, these indicators that were more profitable than any of them individually.

A number of alarm bells started going off as they examined the portfolios that their networks produced. For example, they noticed that much of the portfolios’ paper profits were coming from microcaps—stocks with tiny market caps. That’s troublesome because so few shares of these stocks trade that it’s difficult to establish a sizable position in them without causing their prices to skyrocket. It’s also difficult to borrow shares of these stocks when you want to sell them short.

This heavy reliance on microcaps is just one way in which the AI strategies often make unrealistic assumptions about the real world. Upon restricting their AI strategies to stocks that were relatively easy and cheap to trade, Prof. Avramov and his colleagues found that more than half of those strategies’ paper profits disappeared. And that was before transaction costs, which could easily eat up the remainder of those strategies’ theoretical profits—given that machine learning generates much higher trading volume than that of conventional strategies such as momentum and value investing, according to Prof. Avramov.

Mediocre machines
A perhaps even more surprising conclusion from the study is that the portfolios the AI strategies produced weren’t particularly distinctive. On the contrary, Prof. Avramov says, they were quite similar to portfolios produced by the well-known factors. In other words, these machine-learning techniques largely failed to live up to their promise of finding previously hidden patterns in the stock market.

All in all, it appears that there is “many a slip between the cup and the lip,” to quote the ancient proverb.

This perhaps helps to explain why hedge funds that employ AI haven’t outperformed the S&P 500 over the last decade. Consider the Eurekahedge AI Hedge Fund Index, which “is designed to provide a broad measure of the performance of underlying hedge-fund managers who use artificial intelligence and machine learning theory in their trading processes.” Since its inception in January 2010, the index has produced a 12.7% annualized return, in comparison to a dividend-adjusted 13.3% for the S&P 500.

These results don’t mean that AI is worthless, Prof. Avramov is quick to add. “It’s just that its potential has yet to be proven,” he says. “AI definitely has promise, perhaps not just as much promise as some have made it out to appear.”

Attachment chart.png (23129 bytes) (Download count: 48)


Jan 7, 2020 5:27:13 AM       
Edit 1 times, last edit by ustonapc at Jan 7, 2020 10:21:25 AM
Jrinne
Re: To Quant Or Not To Quant, That Is The Question

Marc,

Does the EurekaHedge AI Hedge Fund Index do better or worse than a similar sample of Designer Models?

I do get that one would want to prove that it is better than the S&P 500.

Renaissance Technologies RIEF fund which is a long term investment for pension funds may address what is possible. It has beaten Ray Dalio’s Bridgewater associate handily over recent years.

Does not prove much as far as what P123 should do I think.

FWIW

-Jim

From time to time you will encounter Luddites, who are beyond redemption.
--de Prado, Marcos López on the topic of machine learning for financial applications

Jan 7, 2020 6:27:00 AM       
ustonapc
Re: To Quant Or Not To Quant, That Is The Question

Jim

Here is the price chart of Renaissance Institutional Equities Funds (RIEF) from Reuters Eikon. Although it has beaten Bridgewater Associates Pure Alpha and ALl Weather Fund in recent years, RIEF seriuosly lags Medallion (which is only open to employees of RT) and despite RIEF's lower volatility still underperform the S&P500. So from a pension fund perspective, it maybe better to invest in the index instead.

Regards
James

Attachment RIEF.png (129643 bytes) (Download count: 37)


Jan 7, 2020 8:01:06 AM       
Jrinne
Re: To Quant Or Not To Quant, That Is The Question

James,

Thank you and so cool!!!!

I have been looking for that data.

-Jim

From time to time you will encounter Luddites, who are beyond redemption.
--de Prado, Marcos López on the topic of machine learning for financial applications

Jan 7, 2020 8:28:47 AM       
Posts: 87    Pages: 9    Prev 1 2 3 4 5 6 7 8 9
 Last Post