Link to publication Citation for published version (APA): Zijlstra, B. J. H. (2017). Regression of directed graphs on independent effects for density and reciprocity. Journal of Mathematical Sociology, 41(4), 185-192. https://doi. ABSTRACT In common models for dyadic network regression, the density and reciprocity parameters are dependent on each other. Here, the j 1 and j 2 models are introduced with a density parameter that represents the log odds of a single tie. Consequently, the density

more »

... reciprocity parameters are independent and the interpretation of both parameters more straightforward. Estimation procedures and simulation results for these new models are discussed as well as an illustrative example.