Duenas-Esterling Marco-Antonio

I'm a 4th year PhD student in Economics at Sant'Anna School of Advanced Studies in Pisa-Italy. My research is related to Trade Networks and Economic Growth. I have a degree in physics, I worked on quantum gates using cold atoms, that was a fantastic theoretical work. Then something happened and I decided to do a MA in Economics, the thesis was related to spatial prisoners dilemma.

To give you an idea about my PhD thesis I investigate whether the gravity model employed in international trade theory can explain the statistical properties of the International Trade Network. The idea is to estimate with econometrics methods the trade flows among countries, or the weights of the links, and then see if these estimations are equivalent to what it is observed in the real network. It can sound strange a "gravity model in economics", but the idea actually works quite nice, the term "gravity" comes about because the predicted relation between trade flows and explanatory variables is similar to Newton's formula: the magnitude of aggregated trade flows between a pair of countries is proportional to the product of country sizes (e.g. the masses, as proxied by country GDPs) and inversely proportional to their geographic distance (interpreted as proxies of trade-resistance factors, e.g. tariffs).

Look that this notion of gravity in social sciences can be adapted to many systems, e.g. migrations, describe how consumers flow between different shopping malls, patients between hospitals... I would be very happy to help interested people in econometric techniques related with the gravity like model, it is connected to Poisson models in count data analysis and Logistic regressions... So if you have the data we could start now to think a project.

One of the conclusions of my research is that the gravity model in trade successfully replicates some weighted-network structure of the trade network, but only if one fixes its binary architecture equal to the observed one. Conversely, the gravity approach performs very badly when asked to predict the presence of a link, or the level of the trade flow it carries, whenever the binary structure must be simultaneously estimated. Hence, one topic I'm very interested is in network formation.

On the side of economic growth I did an empirical analysis of cross sectional GDP business cycles. Business cycles are obtained after applying filtering techniques to country time series, and the cycle is sometimes used in economic literature as white noise that characterize economic fluctuations and volatility. The probability density functions of those cycles show fat-tails, and the volatilities are characterized by an inverse power law relation with the country size. The conclusion is that heterogeneity in economic systems survives to many frequency levels, in other words independently of the filter you can apply there is something that remains, so is the cycle white noise?