Badi H. Baltagi, Peter H. Egger, and Michaela Kesina, "Bayesian Estimation of Multivariate Panel Probits with Higher-Order Network Interdependence and an Application to Firms' Global Market Participation in Guangdong", Journal of Applied Econometrics, forthcoming The data used in the article are information on Chinese firms in the specialized and transport machinery sector in the province of Guangdong. The file bek-data.csv contains the data. It is an unbalanced panel of 1,634 firms observed over three years (2005-2007). The total number of observations is 4,411. The columns contain the following variables: year: 2005-2007 id: individual firm identifier expd: exporter (binary) ford: MNE-network participant (binary) indus3: industry code latitude: latitude (in radians) longitude: longitude (in radians) lemp: labor (in logs) lprod: productivity (in logs) lcapint: capital intensity (in logs) intastr: intangible asset ratio mangfeetowage: management-fee-to-wage-ratio prd: Pearl River Delta (binary) sez: special economic zone (binary) cmp: competition ltastx: fixed export costs (in logs) ltastf: fixed MNE costs (in logs) The file functions.m contains the matlab functions for implementing the proposed estimator. All files are ASCII (text) files in DOS format. They are zipped in the file bek-files.zip. Unix/Linux users should use "unzip -a".