While the positive effect of economic integration on trade is commonly accepted, we still lack a proper understanding of the complex patterns behind this phenomenon. In particular, it is important to better understand how the structure of trade linkages evolves. This is because there are reasons to assume that countries within an economic integration agreement do not trade with each other on random basis. On the contrary, one may argue that they select trade linkages and this choice may be driven by various factors. In this paper we test two specific predictions that originate from the recent literature and which could be informative in this respect. First, we show that the size of the initial trade network is positively correlated with building new trade linkages. In other words, a greater initial number of trading partners facilitates establishing new connections. Second, we also provide some evidence in support of the hypothesis that the evolution of trade network for a given country depends on the trade network of its trading partners. In this case however, our results are slightly less robust.