Does anything beat a GARCH(1,1)? A comparison based on test for superior predictive ability

We compare 330 GARCH-type models in terms of their ability to predict the conditional variance using out-of-sample data. Our question of interest is whether more sophisticated volatility models are able to outperform the simple GARCH(1,1) model. This question is addressed using the test for superior predictive ability (SPA) by [18]. A salient property of this test is that it takes the performance of all models into account simultaneously, thereby avoiding crude approximations and the distortion that arises from pair-wise comparisons. The evaluation is based on daily realized volatility of IBM equity return data and find that the GARCH(1,1) is significantly outperformed by other models, mainly those that accommodate a leverage effect.