Let X, Y and Z be random variables such that S = [[X, Y], [Y, Z]] ~ , where denotes the Wishart distribution and . Define . Then, Var(T) equals
Part BCSIR NET December 2024the-schur-complement-loses-one-degree-of-freedom
The schur complement loses one degree of freedom
Related counterexample: If every marginal is normal then the vector is multivariate normal
- Part B questionJune 2023
- wishart linear formJune 2023
- wishart traceDecember 2023
- wishart quadratic formsDecember 2023
- spearman for bivariate normalDecember 2023
- the conditioning variable determines the shared termDecember 2024
The chapter behind this: Multivariate normal and Wishart — free to read
From Linear Models and Multivariate › Multivariate normal distribution