Do we make better forecasts these days? A survey amongst academics

This paper presents the results of a survey held amongst all editorial board members of six journals. These journals in part focus on the development of models and methods for forecasting. The key question was whether one believes that the forecasting discipline has made progress in the last three decades. Amongst various results, the most important one is that modest progress has been made, although the profession seems far from satisfied. This progress appears to be mainly due to the increase in computing power and the fact that we are better able to incorporate important data features in our models. Additionally, progress could have been faster if we somehow were to include the opinions of experts. These last two findings define two important topics on the research agenda.