A collection of independent identically distributed normal random variables. Orthogonal coordinates in the standardized Gaussian vector separate the sample mean from the residual sum of squares, proving their independence and giving the normal and chi-squared distributions used for exact inference.
New to topics? Read the docs here!