Silverman’s rule of thumb for kernel density estimation bandwidth selection:
$$ h = 0.9 \cdot \min ( \hat{\sigma}, IQR/1.35 ) n^{-1/5} $$Reference
B. W. Silverman “Density Estimation for Statistics and Data Analysis” (1986) // Publisher: Chapman & Hall, London.
Bib
@Article{silverman1986,
title = {Density Estimation for Statistics and Data Analysis},
publisher = {Chapman & Hall, London},
author = {Silverman, B. W.},
year = {1986}
}
Backlinks (1)
- The importance of kernel density estimation bandwidth (2020-10-13) 9 8