The realized kernel (RK) is an estimator of volatility. The estimator is typically computed with high frequency return data, such as second-by-second returns. Unlike the realized variance, the realized kernel is a robust estimator of volatility, in the sense that the realized kernel estimates the appropriate volatility quantity, even when the returns are contaminated with noise. [1]

Notes

edit
  1. ^ Barndorff-Nielsen, Ole E.; Hansen, Peter Reinhard; Lunde, Asger; Shephard, Neil (November 2008). "Designing realised kernels to measure the ex-post variation of equity prices in the presence of noise". Econometrica. 76 (6): 1481–1536. doi:10.3982/ECTA6495. Archived from the original on 2011-07-26.