Title:Space filling in the World Trade Web: measures and null models

Abstract:It is here proposed global and local, bynary and weighted measures of space filling, to detect the degree of stretching of a network within its embedding space. Measures are normalized indexes that vary from zero - a maximally shrunk network, to one -a maximally expanded network. We evaluated the indexes according to three different null models an propose an unbiased measure of network filling, suitable for confrontation of different (cross sectional) embedded networks or historical trends of the same network. We further develop a local measure of filling and thereby of vertex assortativity. Finally, we showed that in the WTW metric distances seem to have determined the links' allocation more than the present null models tend to have predicted and that this burden upon links has been changing in time. However, surprisingly, our analysis seems to indicate that during globalization the WTW expanded merely because of topology and contrary to expectations it actually shrunk.