... using an expansion of 8, (in the case of the k-class estimator) in terms of increasing powers of o, where IJ* is a scalar multiple of the covariance matrix of the errors in the model and the ... of the subvectors p and y: YN = tz;pHz,)-‘z;pfft~l - %&Vh Prv= (r;[p,-p,Z,(Z;p,Z,)-‘Z;p,]y,j-’ ~{Y;[ ~,- ~,~ l~~; ~,~ l~-‘~; ~,] Y,). (3.9) (3.10) In the usual case where H includes Z, ... In a similar way we find that the k-class estimator PCk) of /? has the generic form P(k)= {r;[k(P,-~ ~,) +(l-k,Q ,, ] YZ}-' x{r;[k(P,-P ,, ) +(l-k)Q ,, ] y,} = [kA**(~Z)+(1-k)A22(~)1-‘[ka ,, ( ~,) +(1-k)a ,, ( ~)l....