Research 

Testing for the Interconnection Channel 

• with Bertrand Candelon [Draft available soon]

• Abstract: When modeling large cross-sections in vector autoregressive models, it is typically assumed that each market under analysis depends on few common factors, extracted as linear combination of the variables under analysis. The loadings/weights can be estimated or derived from economic literature, and are usually interpreted as interconnection channels. We propose a novel multiple Likelihood Ratio Test procedure to empirically evaluate the chosen set of weights, and show that such test is fundamental for valid inferences. We exploit the intuition that conditional on the factors employed, no remaining information from the cross section remains statistically significant. The proposed test is intuitive, easy to implement, and presents very good finite sample properties. In the empirical exercise, we test several interconnection channels for the sovereign bond market in the euro area.

• Presentations: 12th International Workshop of Methods in International Finance Network (MIFN), 19th EconomiX-CNRS and Université de Paris Nanterre Econometrics day,  14th Financial Risks International Forum, the 7th RCEA Time Series Workshop, 8th Annual Conference of the International Association for Applied Econometrics (IAAE), 17th Belgian Financial Research Forum, the 27th Annual International Conference on Macroeconomic Analysis and International Finance (ICMAIF).

Modeling Interdependent Assets: A Global Perspective

• with Francesco Roccazzella 

• Abstract: Existing literature has acknowledged two fundamental characteristics of asset returns: they are drifting, and interconnected. We show that it is possible to account for both features jointly, through Global Vector Error Correction modeling. The novel methodology systematically improves the fit of buy-and-hold strategies and, therefore, delivers realistic forecasts of long-term returns, regardless of the asset class under analysis. Moreover, through Generalized Impulse Response analysis we show how to intuitively identify portfolio strategies featuring low exposure to equity market shocks. Interestingly, such strategies consistently outperform traditional approaches mitigating exposure to market fluctuations like minimum variance portfolios or those concentrating on diversified bond portfolios, especially across annual and semi-annual periods.

• Presentations: 2024 International Conference in Finance, Accounting, and Banking, 41st International Conference of the French Finance Association (AFFI), the 29th Annual International Conference on Macroeconomic Analysis and International Finance (ICMAIF).

Robust Bond Risk Premia: Sparse Macroeconomic Information Matters

• with Francesco Roccazzella and Jonas Striaukas 

• Abstract: There is no consensus on whether macroeconomic information significantly improves the prediction of bond risk premia. Existing studies have only examined a limited set of macroeconomic factors, despite the inherently high-dimensional nature of macroeconomic information. We instead propose to test the full set of available macroeconomic variables, assessing the significance of the sparse component in high-dimensional factor-augmented predictive regressions. Three key findings emerge. First, sparse macroeconomic information does matter for bond risk premia prediction. Second, significance emerges consistently from labour market outcomes. Third, the inclusion of labour market variables improves both in- and out-of-sample predictions of bonds excess returns.

The Impact of Wildfires on Loss Given Default: Evidence from Defaulted Consumer Credits

Fragmentation in the European Monetary Union: Is it really over?

European Economic Forecast. Spring 2015.