Pooling versus model selection fornowcasting with many predictors:an application to German GDP

This paper discusses pooling versus model selection for now- and forecasting in the pres-ence of model uncertainty with large, unbalanced datasets. Empirically, unbalanceddata is pervasive in economics and typically due to di¤erent sampling frequencies andpublication delays. Two model classes suited in this context are factor models based onlarge datasets and mixed-data sampling (MIDAS) regressions with few predictors. Thespeci…cation of these models requires several choices related to, amongst others, thefactor estimation method and the number of factors, lag length and indicator selection.Thus, there are many sources of mis-speci…cation when selecting a particular model,and an alternative could be pooling over a large set of models with di¤erent speci…ca-tions. We evaluate the relative performance of pooling and model selection for now-and forecasting quarterly German GDP, a key macroeconomic indicator for the largestcountry in the euro area, with a large set of about one hundred monthly indicators. Ourempirical …ndings provide strong support for pooling over many speci…cations ratherthan selecting a speci…c model.