Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


Thesis formatted SWD.pdf (605.11 KB)

ETD Abstract Container

Abstract Header

Diagnosing Multicollinearity in Exponential Random Graph Models

Duxbury, Scott W
http://rave.ohiolink.edu/etdc/view?acc_num=osu1491393848069144

Abstract Details

2017, Master of Arts, Ohio State University, Sociology.
Exponential random graph models (ERGM) have been widely applied in the social sciences in the past ten years. However, diagnostics for ERGM have lagged behind their use. Collinearity-type problems can emerge without detection when fitting ERGM, skewing coefficients, biasing standard errors, and yielding inconsistent model estimates. This leads to a unique paradox in statistical models of social networks: as more endogenous network effects are modeled, the likelihood of encountering poor model estimates may also increase. This paper provides a method to detect multicollinearity when using ERGM. It outlines the problem and provides a method to estimate shared variance between ERGM parameters. It then tests the method with a Monte Carlo simulation, fitting 27,000 ERGMs and calculating the variance inflation factors for each model. The distribution of variance inflation factors is analyzed using multilevel regression to determine what network characteristics lend themselves to collinearity-type problems. The parameter space of these variables is then examined to specify at what variance inflation factor value a researcher may expect problematic multicollinearity.
Dana Haynie, PhD (Advisor)
David Melamed, PhD (Committee Co-Chair)
Christopher Browning, PhD (Committee Member)
42 p.
Social Research; Social Structure; Sociology; Statistics
Multicollinearity, exponential random graph models, social networks, methodology

Recommended Citations

Citations

  • Duxbury, S. W. (2017). Diagnosing Multicollinearity in Exponential Random Graph Models [Master's thesis, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1491393848069144

    APA Style (7th edition)

  • Duxbury, Scott. Diagnosing Multicollinearity in Exponential Random Graph Models. 2017. Ohio State University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1491393848069144.

    MLA Style (8th edition)

  • Duxbury, Scott. "Diagnosing Multicollinearity in Exponential Random Graph Models." Master's thesis, Ohio State University, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=osu1491393848069144

    Chicago Manual of Style (17th edition)

osu1491393848069144
1,812
© 2017, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.