The Setting
Comparison with a Gaussian cell singles out its centroid as a base direction. The residual scores describe the components of the competing centroids transverse to that direction.
Let have its usual Euclidean inner product and let be standard Gaussian measure on this space. The unnormalized centroid of a measurable cell is the vector integral in (1): (1) For an optimal Gaussian partition , write and for its value.
The Claim
Definition 1 (Negative projection coefficient and orthogonal residual). For vectors , the negative projection coefficient is the real scalar defined by (2), and the orthogonal residual is the vector defined by (3): (2)✦ (3)✦
These residuals are unnormalized; for an optimal partition of value , the corresponding normalized residual is .
Theorem 1 (Residual identities). Let be nonzero and let . The residuals satisfy (4), (5), and (6): (4)✦ (5)✦ (6)✦ For a finite collection with , and an index with , the coefficients satisfy (7): (7)✦ For a finite family of vectors with zero sum and a chosen nonzero vector, the negative projection coefficients of the other vectors relative to the chosen vector sum to one.
In the paper, the result reads: for an optimal Gaussian partition of value , where , having the smallest number of active cells among optimal partitions, order as and put , , , , and for ; then , , , and the Gaussian scores for are independent of , with , , and .
What Is Checked
The statements concern vectors in finite-dimensional real Euclidean space, with a nonzero base vector for the residual identities and a finite collection summing to zero for the coefficient sum identity.