Research
Working papers
Time-varying kernel densities as dynamic infinite mixture models.
Building on kernel density estimation for time series data, we introduce the family of Dynamic Infinite Mixture Models (DIMMs). DIMMs approximate the dynamic density function of a time series with an infinite mixture of location-scale densities. A general specification is presented and several specific models are studied in details. We propose an M-estimator for the fixed parameters and derive its asymptotic properties under a misspecified setting. A consistent estimator for DIMMs-based densities is also introduced. We show that DIMMs outperform extant approaches to kernel density estimation for time series data when it comes to tracking and predicting the time-varying density function of GDP growth. Forecasting results are on par with -- if not better than -- those of a fully fledged parametric model.
Available on SSRN.Economic vulnerability is state dependent (with Leopoldo Catania and Alessandra Luati).
This paper studies the impact that different levels of stress in the financial system have on the real economy. The analysis shows that worsened financial conditions imply: i) a more pessimistic economic outlook when the financial scenario is already severely distressed and, ii) an overall increased macroeconomic uncertainty. Additionally, past information on GDP growth is found to be critical in studying and predicting economic vulnerability. These findings remain valid when other measures of the real economic activity are considered. The study relies on a new methodology for the dynamic modelling of multiple quantiles in the presence of an explanatory variable. Model parameters are estimated through a two-step M-estimation approach. We derive consistency of the resulting estimator and analyse its finite sample properties.
Available on SSRN.A new class of order-preserving lienar trasnformations with an application to modelling multiple quantiles (with Leopoldo Catania and Alessandra Luati).
We introduce a new class of order-preserving linear transformations. These operators map vectors whose entries are sorted in ascending order into equally ordered ones. Starting from this result, we develop a new model to track the quantiles of a time series given all past information on itself and on a set of explanatory variables. This model preserves monotonicity of the estimated quantiles over time. An empirical analysis shows that the new method effectively captures the relation between macroeconomic and financial tail risk. The model also delivers competitive density and tail risk predictions.
First draft to appear soon.New Tests and Estimators for Common Dynamic Factors (with Federico Carlini and Mirco Rubin).
We propose a new estimator for the number of q dynamic factors in a large dimensional dynamic factor model. The estimator is based on a sequential testing procedure for the rank of the residuals’ covariance matrix of the VAR model estimated on Principal Component estimators of the r static factors obtained from a large panel of observations. The rank of the VAR residuals’ covariance matrix is tested by deriving the asymptotically Gaussian distributions of its smallest r-q eigenvalues. We develop both the plug-in and bootstrap version of our eigenvalue-based test. The eigenvectors associated to the q largest eigenvalues allow us to construct an easy-to-implement estimator of the common dynamic factors from the PCs, and to derive its asymptotic properties.
First draft to appear soon.Commonalities in large panels of option prices (with Maria Grith, Paolo Santucci de Magistris and Francesco Violante).
We use a multivariate functional principal component analysis to study commonalities in the implied volatility surfaces of the cross section of US equities. We find that technology-related and financial firms exhibit a higher than average implied volatility. The former accrue this exceedance during the Dot-com bubble, while the latter are more exposed to the Great Financial Crisis. Both findings hold true across different times to maturity and levels of moneyness.
First draft to appear soon.