The Econometrics of Macroeconomic Forecasting
探讨当计量模型与数据生成机制不一致时，如何改进经济预测，涵盖因果信息、简约性、共线性、误差分类、差分与截距校正、协同断裂以及领先指标等概念。
When an econometric model coincides with the mechanism generating the data in an unchanging world, the theory of economic forecasting is reasonably well developed. However, less is known about forecasting when model and mechanism differ in a non‐stationary and changing world. The paper addresses the basic concepts; the invariance of forecast accuracy measures to isomorphic model representations; the roles of causal information, parsimony and collinearity; a reformulated taxonomy of forecast errors; differencing and intercept corrections to robustify forecasts against biases due to shifts in deterministic factors; the removal of structural breaks by co‐breaking; and forecasting using leading indicators.