（数据与代码）Spectral factor modelsWe represent risk factors as sums of orthogonal components capturing fluctuations withcycles of different length. The representation leads to novel spectral factor models in whichsystematic risk is allowed—without being forced—to vary across frequencies. Frequencyspecific systematic risk is captured by a notion of spectral beta. We show that traditionalfactor models restrict the spectral betas to be constant across frequencies. The restrictioncan hide horizon-specific pricing effects that spectral factor models are designed to reveal. We illustrate how the methods may lead to economically meaningful dimensionalityreduction in the factor space.