Serving the Quantitative Finance Community

 
User avatar
FaridMoussaoui
Posts: 327
Joined: June 20th, 2008, 10:05 am
Location: Genève, Genf, Ginevra, Geneva

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

December 18th, 2019, 3:21 pm

The WSJ article is under a paywall. To read it, google the title and click on "I'm feeling lucky", the paywall disappears.....
 
User avatar
tags
Topic Author
Posts: 3162
Joined: February 21st, 2010, 12:58 pm

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

January 5th, 2020, 9:02 am

So, I haven't read the book, but here they argue one takeaway is that Renaissance team had no previous financial knowledge, hence avoiding potential pitfalls.
What do you guys think of this "no prior business knowledge turned into an strength" argument? 
 
 
User avatar
trackstar
Posts: 3420
Joined: January 1st, 1970, 12:00 am

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

January 5th, 2020, 2:20 pm


So, I haven't read the book, but here they argue one takeaway is that Renaissance team had no previous financial knowledge, hence avoiding potential pitfalls.
What do you guys think of this "no prior business knowledge turned into an strength" argument? 
It makes for good dramatic reading, but not likely to be 100% accurate. 

Would you want to trade commodities, for example, through a team that had no knowledge of them?

Lots of arcana there, as you know.

Starting with the simple question: What kind of quantity should I buy 1, 100, 1,000...? 

Opps! 


So the flying blind idea is interesting, but someone has to know enough to do a reality check.

**

Also from your article: 

"Have a single trading model — Instead of creating a unique trading model for every asset class, Henry Laufer felt that one big model for all asset classes would let them leverage the huge swathes of data they had collected and model correlations between different asset classes. This allows future ideas to be added on easily as the model already has an implicit understanding of markets and prices moved. Even for asset classes with smaller history, they could be traded if they were similar to investments with richer history. "

A Great Unified Trading Theory?

Even so, the book will be fun to read.
 
User avatar
tags
Topic Author
Posts: 3162
Joined: February 21st, 2010, 12:58 pm

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

January 5th, 2020, 4:44 pm

Say the 3*** veteran chef makes mayonnaise, taking benefit of skills (s)he has developed during 4 decades or so, and based on an ancestral recipe embedding centuries of cuisine knowledge and know-how. Then, the undergrad in chemistry who has never even cooked makes a bet with friends and decides he will make the best mayonnaise ever using only her/his chemistry class 101 knowledge.
 
User avatar
trackstar
Posts: 3420
Joined: January 1st, 1970, 12:00 am

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

January 5th, 2020, 5:02 pm

Say the 3*** veteran chef makes mayonnaise, taking benefit of skills (s)he has developed during 4 decades or so, and based on an ancestral recipe embedding centuries of cuisine knowledge and know-how. Then, the undergrad in chemistry who has never even cooked makes a bet with friends and decides he will make the best mayonnaise ever using only her/his chemistry class 101 knowledge.
It's an interesting analogy, but evaluating the results too subjective - "chacun a son gout..."

The proof instead is in the P&L pudding...

One of the things that comes up in the book and in discussions elsewhere, is information geometry.  So, if you are feeling adventurous, you might read a few papers on that.  This one was cited on another forum and was published quite recently too - online in July 2019:

Information geometry of modal linear regression - Information Geometry, Springer, (in print June 2019)

Naturally also could read about signal processing and the much broader and intriguing field of information theory. Google Shannon, Gibbs, Boltzmann, and Landauer, for example.

More to the point on finance: one person suggested to look at papers written by RenTech guys before they joined the firm.

A bit of research to dredge those up would be needed, but perhaps worth the effort.
 
User avatar
Alan
Posts: 2958
Joined: December 19th, 2001, 4:01 am
Location: California
Contact:

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

January 5th, 2020, 6:09 pm


So, I haven't read the book, but here they argue one takeaway is that Renaissance team had no previous financial knowledge, hence avoiding potential pitfalls.
What do you guys think of this "no prior business knowledge turned into an strength" argument? 
It makes for good dramatic reading, but not likely to be 100% accurate. 

Would you want to trade commodities, for example, through a team that had no knowledge of them?

Lots of arcana there, as you know.

I've finished the book and, yes, trackstar is right -- the story is a little more complicated about the rationale for "no prior knowledge".

First, they clearly wanted to hire very smart people because, while you can teach smart people the required domain knowledge, you can't teach dumb people to be smart. Simons was experienced in finding such people, and they were often in (non-financial) academia or non-financial research.

Second, the main rationale was a judgement by the seniors that people with no Wall Street experience would be less likely to jump ship for another financial firm.

I thought the book was a very interesting read for the broad story and background stuff like this. But, if you are hoping for any kind of serious discussion of how they made their zillions, you won't find it.
 
 
User avatar
tw
Posts: 592
Joined: May 10th, 2002, 3:30 pm

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

January 9th, 2020, 2:31 pm


I found daveangel's comments on the subject interesting, especially around the relative performance of funds open and closed to outsider investors.
 
User avatar
Alan
Posts: 2958
Joined: December 19th, 2001, 4:01 am
Location: California
Contact:

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

January 9th, 2020, 8:08 pm


I found daveangel's comments on the subject interesting, especially around the relative performance of funds open and closed to outsider investors.

From Bloomberg: The Unsolved Mystery of the Medallion Fund’s Success
 
User avatar
bearish
Posts: 5186
Joined: February 3rd, 2011, 2:19 pm

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

January 10th, 2020, 1:54 am

It feels like they found a leak in the plumbing of the market, and figured out where to put the bucket to catch the money. Then, it's just a matter of concocting a cover story, and making millions of trades a year seems like a good one. I actually doubt that the "alpha" source has anything to do with what most of us would think of as quant investing. But, needless to say, I could be wrong.
 
User avatar
Alan
Posts: 2958
Joined: December 19th, 2001, 4:01 am
Location: California
Contact:

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

January 10th, 2020, 7:51 pm

It feels like they found a leak in the plumbing of the market, and figured out where to put the bucket to catch the money. Then, it's just a matter of concocting a cover story, and making millions of trades a year seems like a good one. I actually doubt that the "alpha" source has anything to do with what most of us would think of as quant investing. But, needless to say, I could be wrong.


That's very interesting. Indeed, five or ten years ago, I thought that Rentech was probably just a very early high frequency trader and they were the beneficiary of the early apparent windfall profits in that space. Unfortunately, nothing later seemed to confirm this, including the book being discussed here. But, I like your idea that the "statistical arbitrage" story is a smokescreen for something else.     
 
User avatar
tw
Posts: 592
Joined: May 10th, 2002, 3:30 pm

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

January 10th, 2020, 11:05 pm

It feels like they found a leak in the plumbing of the market, and figured out where to put the bucket to catch the money. Then, it's just a matter of concocting a cover story, and making millions of trades a year seems like a good one. I actually doubt that the "alpha" source has anything to do with what most of us would think of as quant investing. But, needless to say, I could be wrong.


That's very interesting. Indeed, five or ten years ago, I thought that Rentech was probably just a very early high frequency trader and they were the beneficiary of the early apparent windfall profits in that space. Unfortunately, nothing later seemed to confirm this, including the book being discussed here. But, I like your idea that the "statistical arbitrage" story is a smokescreen for something else.     
How do these high frequency guys with multiple subfunds with mixed internal/external ownership participation and who want 
to remain as secretive about their portfolio composition as possible, avoid the obvious
conflict of interest concerns? Would an auditor provide any assurances of best execution throughout all the subfinds?
 
User avatar
Alan
Posts: 2958
Joined: December 19th, 2001, 4:01 am
Location: California
Contact:

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

January 10th, 2020, 11:28 pm


How do these high frequency guys with multiple subfunds with mixed internal/external ownership participation and who want 
to remain as secretive about their portfolio composition as possible, avoid the obvious
conflict of interest concerns? Would an auditor provide any assurances of best execution throughout all the subfinds?
Probably not a major issue, as the external fund wasn't supposed to resemble the internal one -- see the attached 2 page excerpt.
On the perhaps rare occasion where both funds wanted to buy the same security at the same time, I suppose some proportional allocation rule could be enforced. The auditor might check, and so might the SEC on occasional inspections, to see if the stated policy was being followed.  By the time any outside investor saw a portfolio snapshot of the RIEF fund (if ever), it sounds to me from the book that any putative same ticker purchase would be long gone from the Medallion Fund.  
Attachments
Zuckerman.Excerpt.pdf
(36.05 KiB) Downloaded 175 times
 
User avatar
tw
Posts: 592
Joined: May 10th, 2002, 3:30 pm

Re: The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution

January 11th, 2020, 12:54 am


How do these high frequency guys with multiple subfunds with mixed internal/external ownership participation and who want 
to remain as secretive about their portfolio composition as possible, avoid the obvious
conflict of interest concerns? Would an auditor provide any assurances of best execution throughout all the subfinds?
Probably not a major issue, as the external fund wasn't supposed to resemble the internal one -- see the attached 2 page excerpt.
On the perhaps rare occasion where both funds wanted to buy the same security at the same time, I suppose some proportional allocation rule could be enforced. The auditor might check, and so might the SEC on occasional inspections, to see if the stated policy was being followed.  By the time any outside investor saw a portfolio snapshot of the RIEF fund (if ever), it sounds to me from the book that any putative same ticker purchase would be long gone from the Medallion Fund.  
Thanks! Despite being in this thread, the comment wasn't particularly aimed at Rentech/Simon's funds.

Does the SEC get to see the internal processes? I thought it was all low frequency position reporting (Form 13F etc.)
I just wonder what processes might flag up really simple scams like:  (buy in morning; if market rallies book into this fund, else book into this other fund).
This used to be not uncommon practice for brokers in the bad old days.
I wonder if someone audits timestamps and broker accounts,