***** To join INSNA, visit http://www.insna.org *****
This is an important issue and one that is only recently getting more
focused attention. I think robust identification strategies are an
important area for us to focus on in our current work. Some of these
papers may help. I would also add that "natural experiments" are another
potentially fruitful avenue for addressing this issue. Although natural
experiments are not addressed by the papers below I am currently
conducting several studies in different settings that try to address
identification issues in network research through the use of natural
experiments. So, if you want to discuss that further offline, just email me.
By the way, I would love to hear opinions on the identification issue
from other network scholars on Socnet. Having spoken to several people
about this in various circles, I know there are mixed opinions as to how
critical this issue is and what the best ways to address it are... (you
can put me in the camp that sees it as critical, and the camp that is
actively searching for various solutions:).
Maybe we should start some sort of interest group on this specific issue
if there is enough sustained interest? Not sure how it would run, but I
would love to talk more in depth about how we can address simultaneity
§ Bramoulle, Y., Djebbari, H., and Fortin, B. (2007).
Identification of peer effects through social networks. /Bonn Institute
for the Study of Labor Working Paper/.
§ Kelejian, H.H., and Prucha, I.R. (1998). A generalized spatial
two-stage least square procedure for estimating a spatial autoregressive
model with autoregressive disturbance. /Journal of Real Estate Finance
and Economics/ 104: 219-257.
§ Lee, L.F. (2003). Best spatial two stage least squares
estimators for a spatial autoregressive model with autoregressive
disturbances. /Econometric Reviews/ 22: 307-335.
§ Lee, L.F. (2006). Identification and estimation of econometric
models with group interactions, contextual factors and fixed effects.
/Mimeo// Ohio State University/.
§ Manski, C.F. (1993). Identification of endogenous social
effects: the reflection problem. /Review of Economic Studies/ 60: 531-542.
§ Moffitt, R.A. (2001). Policy interventions, low-level
equilibira, and social interactions, in S. Durlauf and H.P. Young, eds.,
_Social Dynamics_, MIT Press, Cambridge, MA: 45-82.
§ Steglich, C., Snijders, T.A.B., Pearson, M. (2004). Dynamic
networks and behavior: Separating selection from influence. /Mimeo
University of Groningen, The Netherlands/.
Assistant Professor, NYU Stern School of Business.
Research Affiliate, MIT Sloan School of Business.
Personal Webpage: http://web.mit.edu/sinana/www
SSRN Page: http://ssrn.com/author=110270
Corey Phelps wrote:
> ***** To join INSNA, visit http://www.insna.org *****
> I am looking for research papers that explicitly address the (potential)
> endogeneity of node-level network variables (e.g., network size, ego
> density, centrality) in examining the effect of network variables on
> node-level performance. I am interested in efforts to mathematically
> identify the paramter estimates of network variables in such studies using,
> e.g., two-stage estimations with first stage instruments for the network
> variable(s). I am also interested in theoretical treatments of this issue.
> In particular, I would be interested in formal structural models that allow
> one to establish reduced form models for statistical estimation. I will post
> a summary of the responses for the list. Thanks in advance.
> Corey Phelps, PhD
> Asst. Professor, Management & Organization
> University of Washington Business School
> Box 353200
> Seattle, WA 98195
> (206) 543-6579
> SOCNET is a service of INSNA, the professional association for social
> network researchers (http://www.insna.org). To unsubscribe, send
> an email message to [log in to unmask] containing the line
> UNSUBSCRIBE SOCNET in the body of the message.
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.