The value of hard and soft data for short-term forecasting of GDP
通过伪实时预测实验，评估了硬数据（如工业生产）和软数据（如PMI调查）在欧元区GDP短期预测中的价值，揭示了信息可靠性与及时性之间的权衡。
When monitoring and assessing the state of the economy in real time, policymakers face the problem that Gross Domestic Product (GDP) is released with a lag. For the euro area, the first estimate of GDP for a reference quarter is only released six weeks after the close of the quarter. In the interim period, one can use monthly conjunctural indicators to obtain a more timely estimate of GDP. These indicators include hard data, such as industrial production, and soft data such as PMI surveys. However, the hard data for a reference month are only released with a one or two month lag, whereas the soft data are released at the end of the reference month. Hence, one faces a potential trade-off between reliability and timeliness of information. This letter illustrates the value of soft and hard data for computing an early GDP estimate by running a pseudo real-time forecasting exercise.