Chunk #131 — Discussion — Mathematical Details — A moments estimator.

Source

Population structure and eigenanalysis.

Embedded

yes

Text

Let Then So Each entry of M is standard normal, and so T has mean 1 and standard deviation Let s = σ(m,n) be the scale factor of the Johnstone normalization. Then we can show (we used Maple) that as n → ∞, Write T = 1 + x so that x has mean 0 and standard deviation u. Thus, x/s → 0 in probability as m → ∞.