How is generalized sample variance defined?

1.6k Views Asked by At

I'm well aware on this question on Generalized Variance that says that

Generalized variance is the determinant of covariance matrix.

So it would be:

$$\operatorname{var}_G(\matrix{A}) = \operatorname{det}\operatorname{Cov}(\matrix{A})$$

On Wikipedia however, it says:

Another natural generalization of variance for such vector-valued random variables $\matrix{X}$, which results in a scalar value rather than in a matrix, is obtained by interpreting the deviation between the random variable and its mean as the Euclidean distance. This results in $\operatorname{E} \left[(X-\mu)^{\operatorname{T}}(X-\mu )\right]=\operatorname{tr}(C)$, which is the trace of the covariance matrix.

So it would be the sum of the variances of the single vectors in the covariance matrix:

$$\operatorname{var}_G(\matrix{A}) = \operatorname{tr}\operatorname{Cov}(\matrix{A})$$

I'm failing to find a publication defining it on Google Scholar, so I'm asking here how is the generalized sample variance defined?

Thanks!