eduzhai > Applied Sciences > Computer Science >

Weak Identification and Estimation of Social Interaction Models

  • KanKan
  • (0) Download
  • 20210425
  • Save

... pages left unread,continue reading

Document pages: 44 pages

Abstract: The identification of the network effect is based on either group sizevariation, the structure of the network or the relative position in thenetwork. I provide easy-to-verify necessary conditions for identification ofundirected network models based on the number of distinct eigenvalues of theadjacency matrix. Identification of network effects is possible; although inmany empirical situations existing identification strategies may require theuse of many instruments or instruments that could be strongly correlated witheach other. The use of highly correlated instruments or many instruments maylead to weak identification or many instruments bias. This paper proposesregularized versions of the two-stage least squares (2SLS) estimators as asolution to these problems. The proposed estimators are consistent andasymptotically normal. A Monte Carlo study illustrates the properties of theregularized estimators. An empirical application, assessing a local governmenttax competition model, shows the empirical relevance of using regularizationmethods.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×