 Approximate least squares estimation for spatial autoregressive models with covariatesYingying Ma, Wei Lan, Fanying Zhou and Hansheng Wang Computational Statistics & Data Analysis, 2020, vol. 143, issue C Abstract: Due to the rapid development of social network sites, the spatial autoregressive model with covariates has been popularly applied in real practice. However, traditional estimation methods such as the quasi-maximum likelihood estimator are computationally infeasible if the network size n is huge. To circumvent this infeasibility, a novel method named approximate least square estimator (ALSE) is proposed by optimizing an approximate least squares objective function. It can reduce the computational complexity from O(n3) to O(n). Under certain appropriate conditions, the ALSE is consistent and asymptotically normal. In addition, a novel test statistic is proposed to test the identifiability of the parameters in covariates. Extensive simulation studies and a Sina Weibo dataset are analyzed to assess the finite-sample performance of the ALSE. Keywords: Approximate least squares estimator; Quasi-maximum likelihood estimator; Social network data; Spatial autoregressive model (search for similar items in EconPapers) Downloads: (external link) Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:eee:csdana:v:143:y:2020:i:c:s0167947319301884 DOI: 10.1016/j.csda.2019.106833 Access Statistics for this article Computational Statistics & Data Analysis is currently edited by S.P. Azen More articles in Computational Statistics & Data Analysis from Elsevier |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.