In multivariate statistics, if ϵ{\displaystyle \epsilon } is a vector of n{\displaystyle n} random variables, and Λ{\displaystyle \Lambda } is an n{\displaystyle n}-dimensional symmetric matrix, then the scalar quantity ϵTΛϵ{\displaystyle \epsilon ^{T}\Lambda \epsilon } is known as a quadratic form in ϵ{\displaystyle \epsilon }.