Repeated Measures ANOVA using PROC GLM but estimate statements give results by day and group

Question

In SAS I am performing a repeated measures ANOVA of 8 different treatment groups that are measured (continuous) for 9 different times. The log shows no error code only "NOTE: The Huynh-Feldt epsilon and the corresponding adjusted p-value have been enhanced to include a correction based on Lecoutre (1991). Use the UEPSDEF=HF option on the REPEATED statement to revert to the previous definition." To compare the groups I created estimate statements and ran the code below. However, the results are giving my the comparisons of the groups by each day and not overall. Does anyone know how to fix my code to give me overall comparisons of the treatment2 groups instead of by day?

class Treatment2;
model Day0 Day3 Day7 Day9 Day11 Day14 Day16 Day18 Day21 = treatment2;
LSMEANS treatment2; 
REPEATED Day 9;
Estimate "Vehicle vs Axit" intercept 0 treatment2 1 -1 0 0 0 0 0 0;
Estimate "Vehicle vs X4P" intercept 0 treatment2 1 0 -1 0 0 0 0 0;
Estimate "Vehicle vs EMU" intercept 0 treatment2 1 0 0 -1 0 0 0 0;
Estimate "Vehicle vs Axit+X4P" intercept 0 treatment2 1 0 0 0 -1 0 0 0;
Estimate "Vehicle vs Axit+EMU30" intercept 0 treatment2 1 0 0 0 0 -1 0 0;
Estimate "Vehicle vs Axit+EMU10" intercept 0 treatment2 1 0 0 0 0 0 -1 0;
Estimate "Vehicle vs Axit+EMU3" intercept 0 treatment2 1 0 0 0 0 0 0 -1; 
Estimate "Axit vs X4P" intercept 0 treatment2 0 1 -1 0 0 0 0 0;
Estimate "Axit vs EMU" intercept 0 treatment2 0 1 0 -1 0 0 0 0;
Estimate "Axit vs Axit+X4P" intercept 0 treatment2 0 1 0 0 -1 0 0 0;
Estimate "Axit vs Axit+EMU30" intercept 0 treatment2 0 1 0 0 0 -1 0 0;
Estimate "Axit vs Axit+EMU10" intercept 0 treatment2 0 1 0 0 0 0 -1 0;
Estimate "Axit vs Axit+EMU3" intercept 0 treatment2 0 1 0 0 0 0 0 -1; 
Estimate "X4P vs EMU" intercept 0 treatment2 0 0 1 -1 0 0 0 0;
Estimate "X4P vs Axit+X4P" intercept 0 treatment2 0 0 1 0 -1 0 0 0;
Estimate "X4P vs Axit+EMU30" intercept 0 treatment2 0 0 1 0 0 -1 0 0;
Estimate "X4P vs Axit+EMU10" intercept 0 treatment2 0 0 1 0 0 0 -1 0;
Estimate "X4P vs Axit+EMU3" intercept 0 treatment2 0 0 1 0 0 0 0 -1; 
Estimate "EMU vs Axit+X4P" intercept 0 treatment2 0 0 0 1 -1 0 0 0;
Estimate "EMU vs Axit+EMU30" intercept 0 treatment2 0 0 0 1 0 -1 0 0;
Estimate "EMU vs Axit+EMU10" intercept 0 treatment2 0 0 0 1 0 0 -1 0;
Estimate "EMU vs Axit+EMU3" intercept 0 treatment2 0 0 0 1 0 0 0 -1; 
Estimate "Axit+X4P vs Axit+EMU30" intercept 0 treatment2 0 0 0 0 1 -1 0 0;
Estimate "Axit+X4P vs Axit+EMU10" intercept 0 treatment2 0 0 0 0 1 0 -1 0;
Estimate "Axit+X4P vs Axit+EMU3" intercept 0 treatment2 0 0 0 0 1 0 0 -1;
Estimate "Axit+EMU30 vs Axit+EMU10" intercept 0 treatment2 0 0 0 0 0 1 -1 0;
Estimate "Axit+EMU30 vs Axit+EMU3" intercept 0 treatment2 0 0 0 0 0 1 0 -1;
Estimate "Axit+EMU10 vs Axit+EMU3" intercept 0 treatment2 0 0 0 0 0 0 1 -1;
RUN;

Here is some of the data as text

Treatment2  Day0    Day3    Day7    Day9    Day11   Day14   Day16   Day18   Day21
Vehicle 373.21  447.76  470.36  597.19  622.38  660.99  680.88  701.37  709.52
Vehicle 334.65  392.35  425.05  444.53  468.17  501.61  535.23  561.32  586.45
Vehicle 325.69  434.49  486.13  561.87  589.39  617.16  635.58  650.48  672.06
Vehicle 314.07  385.13  409.83  450.49  498.05  535.73  550.16  572.32  593.45
Vehicle 301.95  381.18  407.52  429.94  447.45  475.07  545.13  563.60  579.82
Vehicle 293.52  361.75  427.56  466.02  502.88  525.19  534.77  557.55  569.17
Vehicle 288.34  339.51  386.47  410.21  427.82  444.94  479.95  534.53  553.51
Vehicle 281.00  354.86  389.02  404.25  426.33  460.13  488.96  509.84  523.44
Vehicle 271.92  323.31  394.31  446.21  471.42  505.43  522.66  560.71  584.15
Vehicle 247.63  334.06  374.42  394.74  416.95  435.56  449.11  469.67  502.22
Axitinib    372.35  394.79  457.40  503.90  560.70  584.37  604.30  613.87  611.12
Axitinib    346.61  352.21  399.38  455.42  487.07  514.18  531.80  546.29  495.18
Axitinib    325.22  362.48  421.57  447.87  484.28  494.19  514.91  529.56  545.21
Axitinib    312.67  318.87  349.45  347.28  368.59  351.34  328.07  330.02  340.99
Axitinib    301.26  331.81  353.97  396.04  421.54  423.66  377.51  383.95  339.40
Axitinib    293.22  310.12  328.21  352.37  366.01  416.33  466.34  459.06  498.26
Axitinib    288.02  321.07  312.59  333.53  367.24  397.60  418.85  451.04  462.60
Axitinib    280.62  306.58  327.41  329.91  359.26  369.37  397.47  411.51  419.83
Axitinib    269.21  289.68  289.69  293.77  312.33  332.60  330.56  340.05  316.14
Axitinib    261.17  293.37  316.66  330.84  353.15  370.67  394.63  401.68  412.25
X4P-001 361.16  388.43  417.35  423.77  427.66  394.21  377.00  376.32  390.30
X4P-001 344.86  390.41  424.37  427.28  424.27  413.60  374.13  365.04  366.12
X4P-001 333.89  386.90  412.20  422.83  428.05  410.38  366.45  327.15  313.29
X4P-001 312.44  340.65  365.53  365.56  329.45  301.22  258.29  254.25  237.72
X4P-001 299.49  342.79  340.70  341.01  331.49  292.58  243.38  230.48  214.18
X4P-001 292.75  333.85  354.13  355.74  323.65  272.98  182.93  171.76  155.85
X4P-001 285.62  314.95  350.49  356.16  342.83  310.56  282.52  268.74  244.62
X4P-001 280.07  310.08  369.17  386.90  382.47  321.35  293.53  272.28  293.15
X4P-001 268.18  280.82  284.08  277.09  256.50  231.94  213.96  194.19  168.69
X4P-001 260.42  259.00  283.04  269.19  250.15  239.06  184.84  174.79  159.91
EMU-116 357.55  385.40  412.77  413.42  402.65  391.73  384.69  347.43  351.78
EMU-116 342.97  387.83  401.87  429.14  434.01  433.11  402.85  373.37  379.85
EMU-116 333.19  335.16  368.13  360.60  358.52  307.19  278.11  240.87  241.94
EMU-116 324.29  366.84  394.14  391.03  376.36  367.10  353.80  340.94  335.84
EMU-116 299.09  336.93  365.86  374.62  365.07  321.82  298.34  292.71  311.85
EMU-116 292.58  322.19  343.59  341.74  333.74  327.12  303.06  298.77  296.64
EMU-116 285.07  318.62  294.64  287.21  272.97  266.17  274.87  285.30  292.95
EMU-116 278.57  311.09  242.52  224.47  201.47  163.17  140.89  127.43  119.68
EMU-116 267.38  283.47  303.50  308.15  309.08  271.16  227.43  184.96  181.43
EMU-116 260.17  249.17  269.25  256.51  249.21  220.81  197.33  187.98  178.38
Axitinib+X4P-001    353.53  369.35  410.53  412.34  436.85  452.25  438.86  453.10  396.18
Axitinib+X4P-001    341.29  313.73  354.51  344.83  348.97  320.62  328.67  321.49  308.48
Axitinib+X4P-001    332.97  333.99  363.80  365.39  367.08  378.15  341.22  376.28  407.15
Axitinib+X4P-001    322.87  320.00  355.81  334.56  316.77  306.68  288.80  267.51  243.59
Axitinib+X4P-001    311.57  366.84  438.02  419.72  432.72  458.63  469.85  488.41  478.12
Axitinib+X4P-001    292.45  283.93  310.38  316.24  305.01  293.05  259.21  267.79  321.79
Axitinib+X4P-001    284.99  268.87  262.26  243.26  228.36  189.66  179.34  151.56  136.16
Axitinib+X4P-001    276.49  268.14  288.18  274.23  285.68  280.96  301.35  319.24  279.93
Axitinib+X4P-001    266.92  255.73  299.85  296.95  281.90  287.72  291.23  279.59  261.19
Axitinib+X4P-001    259.06  245.11  263.67  269.42  266.59  225.00  227.68  250.29  267.86

How about if you just average the multivariate responses together and then run your model with that average response? — Russ Lenth, Jun 12 '21 at 03:45
Been awhile since I used SAS, but can't you do `lsmeans treat2 / diff` or something like that in lieu of 28 `estimate` statements? — Russ Lenth, Jun 12 '21 at 12:31

score 1 · Answer 1 · answered Jun 12 '21 at 21:01

It's certainly easy enough to do in R:

> SOmod = lm(cbind(Day0, Day3, Day7, Day9, Day11, Day14,
+                  Day16, Day18, Day21) ~ Treatment2, data = SOdat)

> library(emmeans)

> (SOemm = emmeans(SOmod, "Treatment2"))
 Treatment2       emmean   SE df lower.CL upper.CL
 Axitinib            392 19.9 45      352      432
 Axitinib+X4P-001    317 19.9 45      277      357
 EMU-116             308 19.9 45      268      348
 Vehicle             473 19.9 45      433      513
 X4P-001             314 19.9 45      274      354

Results are averaged over the levels of: rep.meas 
Confidence level used: 0.95 

> pairs(SOemm)
 contrast                       estimate   SE df t.ratio p.value
 Axitinib - (Axitinib+X4P-001)     74.77 28.2 45  2.653  0.0777 
 Axitinib - (EMU-116)              84.01 28.2 45  2.981  0.0355 
 Axitinib - Vehicle               -80.97 28.2 45 -2.873  0.0464 
 Axitinib - (X4P-001)              78.13 28.2 45  2.772  0.0590 
 (Axitinib+X4P-001) - (EMU-116)     9.24 28.2 45  0.328  0.9974 
 (Axitinib+X4P-001) - Vehicle    -155.74 28.2 45 -5.525  <.0001 
 (Axitinib+X4P-001) - (X4P-001)     3.36 28.2 45  0.119  1.0000 
 (EMU-116) - Vehicle             -164.98 28.2 45 -5.853  <.0001 
 (EMU-116) - (X4P-001)             -5.88 28.2 45 -0.209  0.9996 
 Vehicle - (X4P-001)              159.10 28.2 45  5.645  <.0001 

Results are averaged over the levels of: rep.meas 
P value adjustment: tukey method for comparing a family of 5 estimates

Repeated Measures ANOVA using PROC GLM but estimate statements give results by day and group

1 Answers1