Andrey Akinshin's Homepage

Selected writings on statistics:

Weighted quantile estimators
Refined Mann–Whitney U Test
Trimmed Harrell-Davis quantile estimator based on the highest density interval of the given wid... · 2022
Quantile-respectful density estimation based on the Harrell-Davis quantile estimator · 2024
Quantile absolute deviation · 2022
Finite-sample Rousseeuw-Croux scale estimators · 2022
Finite-sample bias-correction factors for the median absolute deviation based on the Harrell-Da... · 2022

Selected writings on other topics:

On a Numerical Model of a Circadian Oscillator · 2022
Geometry of error amplification in solving the Prony system with near-colliding nodes · 2021
Accuracy of reconstruction of spike-trains with two near-colliding nodes · 2017
Accuracy of spike-train Fourier reconstruction for colliding nodes · 2015
Pro .NET Benchmarking · 2019

Selected blog posts:

Effect Sizes and Asymmetry · 2024-03-12
The Effect Existence, Its Magnitude, and the Goals · 2024-02-27
Degrees of practical significance · 2024-02-06
Eclectic Statistics
Sporadic noise problem in change point detection · 2023-11-28
Parametric, Nonparametric, Robust, and Defensive statistics · 2023-08-08
Insidious implicit statistical assumptions · 2023-08-01
Confusing tie correction in the classic Mann-Whitney U test implementation · 2023-05-23
Performance stability of GitHub Actions · 2023-03-21
Adaptation of continuous scale measures to discrete distributions · 2023-04-04
p-value distribution of the Mann–Whitney U test in the finite case · 2023-02-28
Debunking the myth about ozone holes, NASA, and outlier removal · 2023-01-31
Trinal statistical thresholds · 2023-01-17
The Huggins-Roy family of effective sample sizes · 2022-09-13
Gastwirth's location estimator · 2022-06-07
Misleading standard deviation · 2021-02-23
Misleading histograms · 2020-10-20
The importance of kernel density estimation bandwidth · 2020-10-13
Implementation of an efficient algorithm for changepoint detection: ED-PELT · 2019-10-07
65535 interfaces ought to be enough for anybody · 2017-02-14

More: