reml

Let, \mathbf y_i = \mathbf X_i\mathbf\beta + \mathbf Z_i\mathbf b_i+ \mathbf\epsilon_i, where \mathbf y_i\sim N(\mathbf X_i\mathbf\beta, \Sigma_i=\sigma^2\mathbf I_{n_i}+\mathbf Z_i \mathbf G\mathbf Z_i'), \mathbf b_i\sim N(\mathbf 0, \mathbf G), \mathbf\epsilon_i\sim N(\mathbf 0...

I have to compute the following differentiation : \frac{\partial}{\partial\sigma^2}\det[\mathbf X_{p\times n}'(\sigma^2 \mathbf I_{n}+\mathbf Z_{n\times q}\mathbf G_{q\times q}\mathbf Z_{q\times n}')^{-1}\mathbf X_{n\times p}], where \sigma^2 is a scalar, \det denotes determinant, \mathbf...

Dear All I am trying to get my head around when to use ANOVA when to use regeression and when to use REML. As I understand REML can be used in situations where you would normally use ANOVA (2 variables?) but for unbalanced / correlated data, or where you would use linear regression but have...