Different variances for each time point

Sometimes variability in a measurement will differ between the time points. This was allowed for in the general covariance pattern (i) above. Some additional patterns allowing for differing variances are given below. In pattern (v) time points have different variances but observations on the same patient are uncorrelated. This should only be used if preliminary analyses with more parameterised patterns indicate a lack of correlation between the repeated observations. Patterns (vi)-(viii) have similar forms to the autoregressive, compound symmetry and Toeplitz patterns shown above, except that different variances for each time point are now used.

(v) Heterogeneous uncorrelated

Ri =

(01

0

0

0

0

022

0

0

0

0

032

0

0

0

042

(vi) Heterogeneous compound symmetry

PCT1CT2 o2 PO1O3 PO2O3

PO1O2 PO1O3

P0104

P0203 P0204

P0304

\p0104 PO2O4 PO3O4 o42 f

(vii) Heterogeneous first-order autoregressive

PO1O2

p 0103

\p30104

PO1O2

P0203

p20103

P0203

o p 0204 P0304

p30104

p 0204

P0304

(viii) Heterogeneous Toeplitz

01 P10102 P20103 P3 0104

2 P10203 P2 0204 02

p1 01 02 02 p20103 p10203

p1 03 04

\p30104 P20204 P10304 042

Was this article helpful?

0 0

Post a comment