Kunyu's Blog

Success comes from standing out,
not fitting in.

在知乎答：为什么样本方差的分母是n - 1？

翻译自之前的博客：Estimate Population Variance: should we divide by n - 1 or n

该回答写在知乎问题为什么样本方差（sample variance）的分母是 n-1？下，现搬运到此。内容与我的英文博客：Estimate Population Variance: should we divide by n - 1 or n 大致相同。 问题下已经有许多非常精彩的回答了。在估计总体方差时，比较常用的估计量（estimator）包括样本方差估计量 $s^{2}=\...

Posted by 贺鲲羽 on February 26, 2021

Estimate Population Variance: should we divide by n - 1 or n

Understand why we use (n − 1) in sample variance, and why dividing by n still gives us a good estimator

This article discusses how we estimate the population variance of a normal distribution, often denoted as $\sigma^2$. Typically, we use the sample variance estimator defined as: \begin{equation}s^...

Posted by Kunyu He on April 19, 2020

Principal Component Analysis (PCA)

Understand PCA and how we do it via EVD and SVD, and why the SVD implementation is better

This article discusses what is principal component analysis (PCA), how we do it using eigenvalue decomposition (EVD) or singular value decomposition (SVD), and why the SVD implementation is better....

Posted by Kunyu He on February 23, 2020

Multicollinearity and Ridge Regression

Understand multicollinearity and how it compromises least squares, and how ridge regression helps

This article discusses what is multicollinearity, how can it compromise least squares, and how ridge regression helps avoid that from a perspective of singular value decomposition (SVD). It is heav...

Posted by Kunyu He on January 11, 2020

SVD and Underdetermined Least Squares

Understand how SVD derives a consistent expression for least-square weights

This blog discusses the difference in least-squares weight vectors across over- and underdetermined linear systems, and how singular value decomposition (SVD) can be applied to derive a consistent ...

Posted by Kunyu He on December 31, 2019

FEATURED TAGS
Machine Learning SVD Bias–variance Tradeoff Linear Regression Statistical Learning
ABOUT ME

Happiness is the moment before you need more happiness.
秋色无南北，人心自深浅

FRIENDS