Skip to content
Get started

The best divisor for a variance estimate

HardStatistics6 min

You have n=9n=9 i.i.d. normal observations with unknown mean and variance. You estimate σ2\sigma^2 by 1c∑i=19(Xi−Xˉ)2\dfrac{1}{c}\sum_{i=1}^9 (X_i-\bar X)^2 and choose the constant cc.

Which integer cc minimises the mean squared error E[(σ^2−σ2)2]E[(\hat\sigma^2-\sigma^2)^2]?

Show hints (2)+
  1. Write SSD=σ2W\text{SSD}=\sigma^2W with WW chi-square on n−1n-1 degrees of freedom.
  2. MSE = variance + bias²; minimise over u=1/cu=1/c.

Answer

Reveal answer →

10

Create a free account to see the full worked solution and track your progress.

Interview format: Hedge Fund Quant, Systematic & Data-Driven Trading

Related questions