***** To join INSNA, visit http://www.insna.org *****
On Mar 29, 2007, at 3:03 PM, Michael Haenlein wrote:
> The question I would like to ask is: Is a customer more likely to
> leave (i.e. resign) when one of his 'friends' left previously?
> Essentially this
> means that I would like to incorporate social network effects into
> survival analysis (e.g. Cox regression). My idea is somewhat to make
> the hazard rate of one customer dependent on the hazard rates of this
> customer's friends, but I'm not really sure how to do that.
Making the hazard rate a function of the network neighbors' hazard rate
is difficult. It is much easier to make the hazard rate a function of
the "state" of the network neighbors. I also think it is quite appealing
from a substantive/intuitive point of view. There are several papers in
the innovation adoption literature doing this. For a general discussion,
Strang, David (1991), "Adding Social Structure to Diffusion Models: An
Event History Framework," Sociological Methods & Research , 19
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.