A multivariate spatial econometrics model with an intra-location feedback effect

Research output: Contribution to conferenceOther

Abstract

In the spatial multivariate econometrics models, the relations are typically built for the between location dependencies, which is adopted from univariate case like SEM and SLM models. However, this does not allow for intra-location dependencies to be accounted directly. The weakness of the previous models is shown analytically and using examples. A new multivariate spatial econometric model is presented that accounts for a feedback effect between variables within the same location, called intra-location feedback effect, which was not considered before in this area. Model identification and other basic properties of the spatial multivariate econometrics model, and especially for the extended version, are established. Statistical inferences are presented using a proposed analytical method based on empirical precision matrix and also the maximum likelihood. Model validation for new data sets is discussed.

Details

Authors
Organisations
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Economics
  • Probability Theory and Statistics
Original languageEnglish
Publication statusPublished - 2016
Publication categoryResearch
Peer-reviewedNo
EventInternational Conference on Computational and Financial Econometrics - Higher Technical School of Engineering of Seville, Seville, Spain
Duration: 2016 Dec 92016 Dec 11
Conference number: 10
http://cfenetwork.org/CFE2016/

Conference

ConferenceInternational Conference on Computational and Financial Econometrics
Abbreviated titleCFE
CountrySpain
CitySeville
Period2016/12/092016/12/11
Internet address