Forecasting the intra-day effective bid ask spread by combining density forecasts

The bid-ask spread refers to the tightness dimension of liquidity and can be used as a proxy for transaction costs. Despite importance in financial literature, few studies have investigated its forecastability. We propose new methodology predict bid ask by combining density forecasts two types models: Multiplicative Errors Models ARMA-GARCH models. Our method is employed effective intra-day series all shares pertaining CAC40 index. Using one-step-ahead out-of-sample framework, we resort on Model Confidence Set procedure classify models found that proposed model appears beat benchmark specifications.