We examine 24 global factor premiums across equity, bond, commodity, and currency markets via replication and out-of-sample evidence between 1800 and 2016. Replication yields ambiguous evidence within a unified testing framework that accounts for p-hacking. Out-of-sample tests reveal strong and robust presence of the large majority of global factor premiums, with limited out-of-sample decay of the premiums. We find global factor premiums to be generally unrelated to market, downside, or macroeconomic risks in the 217 years of data. These results reveal significant global factor premiums that present a challenge to traditional asset pricing theories.

我们通过复制和 1800 年至 2016 年间的样本外证据，研究了股票、债券、商品和货币市场的 24 种全球因子溢价。在一个统一的测试框架内，复制产生了模棱两可的证据，该框架考虑了 p-hacking。样本外测试显示，绝大多数全球因子溢价都存在强劲而稳健的情况，保费的样本外衰减有限。我们发现，在 217 年的数据中，全球要素溢价通常与市场、下行或宏观经济风险无关。这些结果揭示了显着的全球要素溢价，这对传统资产定价理论提出了挑战。