Carry Trade Returns with Support Vector Machines

This paper proposes a novel approach to directional forecasts for carry trade strategies based on support vector machines (SVMs), a learning algorithm that delivers extremely promising results. Building on recent findings in the literature on carry trade, we condition the SVM on indicators of uncertainty and risk. We show that this provides a dramatic performance improvement in strategy, particularly during periods of financial distress such as the recent financial crises. Disentangling the measures of risk, we show that conditioning the SVM on measures of liquidity risk rather than on market volatility yields the best performance.     

University specialization and new firm creation across industries

This article examines how the scientific specialization of universities impacts new firm creation across industries at the local level. In accordance with the Pavitt-Miozzo-Soete taxonomy, we consider eight industry categories, which reflect the characteristics of firms’ innovation patterns and, ultimately, the knowledge inputs that firms require. Using data on new firm creation in Italian provinces (i.e., at the NUTS3 level), we estimate negative binomial regression models separately for each industry category to relate new firm creation to the scientific specialization in basic sciences, applied sciences and engineering, and social sciences and humanities of neighboring universities. We find that universities specialized in applied sciences and engineering have a broad positive effect on new firm creation in a given province, this effect being especially strong in service industries. Conversely, the positive effect of university specialization in basic sciences is confined to new firm creation in science-based manufacturing industries, even if this effect is of large magnitude. Universities specialized in social sciences and humanities have no effect on new firm creation at the local level whatever industry category is considered.