AI 中文总结
本文基于美国2000-2026年数据，对比实时与修正数据下ML模型和传统模型的通胀预测表现，发现ML优势在实时数据下减弱，SHAP特征重要性排名存在事后偏差，RSI可用于审计ML通胀预测。
AI 中文摘要
机器学习（ML）通胀预测几乎普遍基于完全修正后的数据训练，然而实时预测者从未能获得此类数据，且报告的特征重要性通常是通过样本内计算得出的，这混淆了预测相关性与回溯拟合度。本文探究Agyekum（2026）中所记录的ML相对于菲利普斯曲线的优势，在模型基于实时（ALFRED） vintage而非修正序列进行训练和评估时是否仍然存在，以及SHAP特征重要性排名是否为样本内估计的产物。利用2000-2026年美国的失业率、CPI、PCE通胀、就业人数、实际GDP以及10年期-2年期国债利差数据，构建了4种传统模型（随机游走、AR(1)、菲利普斯曲线、ADL-OLS）和4种ML模型（随机森林、梯度提升、弹性网、SVR）的vintage一致面板，在3、6、12个月的预测期内递归重新估计（分别产生208、206、204个预测值）。实时/修正数据的精度差异很小，除了6个月期的一个例外（梯度提升vs菲利普斯曲线，DM=-1.671，p=0.097），在Diebold-Mariano检验下无显著差异；仅梯度提升在较长预测期表现出持续的正预测能力。随机游走仍是强劲的短期基准，与Agyekum等人（2026）关于汇率的谜题一致。采用滚动向前的样本外SHAP，基于修正数据的随机森林将主要重要性赋予PCE通胀（平均|SHAP|=0.778，9个特征中排名第1），而基于实时数据时，其赋予PCE的重要性可忽略不计（0.039，排名第6），转而依赖当期CPI（0.834，排名第1，PCE排名第2为0.223）。这一20倍的波动幅度大于样本内估计值，无法被点预测指标察觉，表明模型对PCE的依赖在很大程度上是一种事后产物。RSI按模型和预测期汇总了精度差距，对ML通胀预测的审计具有启示意义。
英文摘要
ML inflation forecasts are almost universally trained on fully revised data, even though real-time forecasters never have such data, and reported feature importances are typically computed in-sample, conflating predictive relevance with retrospective fit. This paper asks whether the ML advantage over the Phillips curve documented in Agyekum (2026) survives when models are trained and evaluated on real-time (ALFRED) vintages rather than revised series, and whether SHAP feature-importance rankings are an artifact of in-sample estimation. Using 2000-2026 U.S. data on unemployment, CPI and PCE inflation, payrolls, real GDP, and the 10-year-2-year Treasury spread, vintage-consistent panels are built for four traditional models (random walk, AR(1), Phillips curve, ADL-OLS) and four ML models (Random Forest, Gradient Boosting, Elastic Net, SVR), re-estimated recursively at 3-, 6-, and 12-month horizons (208, 206, 204 forecasts). Real-time/revised accuracy differences are small and, apart from one exception at 6 months (Gradient Boosting vs. Phillips curve, DM = -1.671, p = 0.097), indistinguishable under Diebold-Mariano tests; Gradient Boosting alone shows consistent positive skill at longer horizons. The random walk remains a strong short-horizon benchmark, consistent with the puzzle in Agyekum et al. (2026) for exchange rates. Using walk-forward, out-of-sample SHAP, a Random Forest on revised data assigns dominant importance to PCE inflation (mean |SHAP| = 0.778, rank 1 of 9), while on real-time data it assigns PCE negligible importance (0.039, rank 6), relying instead on current CPI (0.834 vs. 0.223). This twenty-fold swing, larger than the in-sample estimate, is invisible to point-forecast metrics and shows the model's PCE reliance is substantially a hindsight artifact. An RSI summarizes the accuracy gap by model and horizon, with implications for auditing ML inflation forecasts.
