Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: multiple -spmat imports- to manually import higher-order links


From   László Sándor <[email protected]>
To   [email protected]
Subject   st: multiple -spmat imports- to manually import higher-order links
Date   Sun, 10 Jul 2011 20:45:10 -0400

Hi all,

I am using Stata 11.2 on unix and mac for some social network analysis
with -sppack-, and I am back with another question after benefitting
from Rafal's previous generous help on this list in the last few
weeks.

I'm basically still stuck with the nature of my problem involving
spatial/social links in a chronological order, not allowing me to let
higher-order links be simply defined by raising the direct adjacency
matrix to the respective power. I think I could code up the tracing
out of the links separately, but for the ultimate ML estimation of the
spatial autoregressive linear regression, I would need to impute a
correlation matrix in the likelihood function with a single AR
parameter driving the exponential dampening of higher-order links,
while I add in the links "manually" from the "adjacency" matrices of
successive orders.

I.e. if I import a neighbor-list of first-order links  in A, of
second-order links in B etc., I can have a parsimonious likelihood
funtion where the correlation of observations is allowed to be I +
rho*A + rho^2*B + ..., where rho is a parameter of the function and
will be estimated.

I think it needs be done with my own ML code, of course, but to get
the right matrices from the link data, it would still be nice to use
something like -spmat import-. So the question for today is whether
the -spmat- object is usable for anything else than use in -spreg- and
-spivreg- for more restrictive spatial AR models? Can I use it to
import the matrices A, B etc. separately for further manipulation?

Thank you very much, I hope these answers benefit the wider community too.

Laszlo
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index