En coll. avec Jean-Laurent Viviani (CREM, Univ. Rennes 1)
Abstract: We compare the forecasting power of a series of univariate econometrics models to predict the effective intra-day bid-ask spread series of CAC40 constituents. Multiplicative Errors Models and ARMA-GARCH types of models are compared in a one-step-ahead out-of-sample density forecast framework. The Model Confidence Set procedure of Hansen et al. (2011) is employed to classify models and the potential benefit of combining density forecast is investigated.
Keywords: Liquidity, Multiplicative Errors Models, Forecasting