 Extended Generalized Estimating Equations (v0.5)
        for Correlated Binary Data
         (c) Bahjat Qaqish, 1989, 1990, 1991

 Date: 10/05/92    Time: 13:54:37

 -- ECG data / crossover design --
 -- compare EEG1 and EEG2 --
 Data file: ECG04 DATA
 dim(x):            3
 dim(b):            4
 # classes          1
 max #iter:        50
 tolerance:  0.0001000000000
 # clusters:       67
 cluster size          #clusters
      2                     67
 The model:
 1 : B1 X1  B2 X2  B3 X3
 1.1 : B4 X1

 Method of computing 3rd and 4th order moments: The exact method.
 -- GEE2 --

 Converged after     3 iterations.
 Method of computing 3rd and 4th order moments: The exact method.
 -- GEE2 --

 Estimates:
 Beta:                    Estimate    Robust s.e.     Naive s.e.     Robust z
 1 %GM                  -1.28514        0.36737        0.37941       -3.49821
 2 Treatment             0.55868        0.23652        0.23792        2.36204
 3 Sequence              0.41650        0.47576        0.47678        0.87544
 4 LogOR:1               3.35596        0.82424        0.82338        4.07159
 (Lower = Correlation,  Upper = Variance)
 Robust Correlation   Variance Matrix Estimate:
 ( 1)  0.13496  -0.03119  -0.11407  -0.00065
 ( 2) -0.35892   0.05594  -0.00572   0.00042
 ( 3) -0.65262  -0.05081   0.22635   0.00073
 ( 4) -0.00215   0.00215   0.00187   0.67937

 Naive  Correlation   Variance Matrix Estimate:
 ( 1)  0.14395  -0.03729  -0.12147   0.00000
 ( 2) -0.41305   0.05661   0.00316  -0.00000
 ( 3) -0.67151   0.02784   0.22732  -0.00000
 ( 4)  0.00000  -0.00000  -0.00000   0.67795

 Date: 10/05/92    Time: 13:54:41
 CPU time:      0.422 seconds =      0.007 minutes.