February 2008

Sun Mon Tue Wed Thu Fri Sat
          1 2
3

4

5

6

7

8

9

10 11 12 13 14 15 16
17 18

19

20 21 22 23

24

25

26

27

28

29  

Editor Login


Convener in chief:


David Lazer
(Methodology, Networked Governance)

Editors:


Stanley Wasserman
(Current Trends, Methodology, Social Networks)

Allan Friedman
(Simulations)

Nathan Eagle
(Technology, Social Computing, Powerlaws, Current Trends)

Ben Waber
(Technology, Social Computing)
Thomas Langenberg
(Technology, Social Computing, Social Networks, Current Trends)

Ines Mergel
(Knowledge Sharing, Social Computing, Social Software, Current Trends)

Brian Rubineau
(Social Dynamics, Societal Networks, Simulations)

Maria Binz-Scharf
(Qualitative Methodology, Knowledge Sharing, eGovernment)

Jeff Boase
(Technology, Societal networks)

Alexander Schellong
(Admin, eGovernment, Citizen Relationship Management)

Categories

Archives

Recent Entries

Recent Comments

Notification


« Friends of my friends are my… enemies? | Main | I am creating my own online social network: Team TriOdy and Social Science Research in Practice »

9 June 2007

The small world of investing

Interesting NBER paper discussed in today’s NYT, by Laren Cohen, Andrea Frazzini, and Christopher Malloy. The gist of it: that investments by mutual funds are more likely in companies in which the manager of a fund has classmates on the board of directors, and, further, the returns on those investments are far higher than investments in which the manager does not have “old school ties.” Further, the timing of these investments is tied to particular news events from the corporation, suggestion some “information transfer” is occurring.

This is both a neat bit of research, which makes a persuasive case for the presence of information transfer. The magnitude of the effect is particularly striking when you consider that their proxies for a social network, while reasonable, are certainly noisy. That is, while attending the same school, especially in the same cohort, is certainly strongly predictive of a relationship, people do not have relationships with all of their classmates, and have many/most of their relationships with non-classmates. So this effect (almost double the return for investments associated with ties as compared to those not) is likely a fairly conservative estimate of the impact of social ties.

I would also note that this paper is also illustrative of entry of social network ideas into economics (generally by a new cohort of scholars), as well as (to revisit my earlier discussions about computational social science) the potential to mine the growing mountains of relational data that now exist in various archives.

Posted by David Lazer at June 9, 2007 9:43 AM

Comments

A line in the piece also tells us

>The most common shared school in the study, by far, was Harvard Business School.

A preponderance for one school leads to the question, did the data allow to control for causality by school (rather than of course by social tie). Likely the paper tells.

Posted by: David Allen at June 10, 2007 1:20 PM

Notification

Enter e-mail address to receive notification of new comments to this entry

Post a comment




Remember Me?