Keywords: social network analysis, dynamic network model, latent factor model
The General Assembly of the United Nations serves as the key component of policymaking of the United Nations. 193 members vote on various of events each year. Nations' foreign policy preferences and positions can be revealed from those voting behaviors. In this paper, we analyze the UN voting data (Voeten 2013) using a dynamic latent factor network model. We introduce an extension of the multiplicative latent factor model (Hoff 2009), by incorporating the time-dependent covariance structure to the prior specifications of the parameters. By controlling for geographic distances and bilateral trade between countries, the model estimated latent positions and movement of positions reveal interesting and meaningful foreign policy positions and alliance of various countries.