y*****w
发帖数: 1350 | 1 If anyone happens to have SAS 9.22 or SAS 9.3 (not SAS 9.2), may I have an
earnest request that you take half a minute to run PROC POWER for me? I am
asking this because I only have 9.2 whereas there is a new option
EVENTSTOTAL= in 9.22/9.3, which is exactly the number I need to know now.
The SAS code is as follows which automatically calculates number of total
events:
proc power;
twosamplesurvival test=logrank
groupsurvexphazards = (0.001684 0.000838)
accrualtime = 0
followuptime = 12
power = 0.80
npergroup = .;
run;
Thank you!! |
n**********e
发帖数: 18 | 2 The POWER Procedure
Log-Rank Test for Two Survival Curves
Fixed Scenario Elements
Method Lakatos normal
approximation
Form of Survival Curve 1
Exponential
Form of Survival Curve 2
Exponential
Accrual Time
0
Follow-up Time
12
Group 1 Survival Exponential Hazard 0.
001684
Group 2 Survival Exponential Hazard 0.
000838
Nominal Power
0.8
Number of Sides
2
Number of Time Sub-Intervals
12
Group 1 Loss Exponential Hazard
0
Group 2 Loss Exponential Hazard
0
Alpha
0.05
Computed N Per Group
Actual N Per
Power Group
0.800 2321 |
n**********e
发帖数: 18 | 3 不能和npergroup一起用，output 如下：
Fixed Scenario Elements
Method Lakatos normal
approximation
Form of Survival Curve 1
Exponential
Form of Survival Curve 2
Exponential
Accrual Time
0
Follow-up Time
12
Group 1 Survival Exponential Hazard 0.
001684
Group 2 Survival Exponential Hazard 0.
000838
Nominal Power
0.8
Number of Sides
2
Number of Time Sub-Intervals
12
Group 1 Loss Exponential Hazard
0
Group 2 Loss Exponential Hazard
0
Alpha
0.05
Group 1 Weight
1
Group 2 Weight
1
Computed Ceiling Event Total
Ceiling
Fractional Actual Event
Event Total Power Total
69.630853 0.846 70 |
n**********e
发帖数: 18 | 4 see the next reply
Computed Ceiling Event Total
Ceiling
Fractional Actual Event
Event Total Power Total
69.630853 0.846 70 |
y*****w
发帖数: 1350 | 5 Thank you neardistance! If npergroup= and eventstotal= cannot appear at the
same time, could you try one more time using the following code? If it could
not run, then add "accrualtime=0". If you have already run the code like
that, then stop here. I am asking this because there is no "Computing Actual
Event Total" in your output, which is also what I would like to know.
Thanks!
proc power;
twosamplesurvival test=logrank
groupsurvexphazards = (0.001684 0.000838)
followuptime = 12
power = 0.80
eventstotal = .;
run;
【在 n**********e 的大作中提到】
: see the next reply
: Computed Ceiling Event Total
: Ceiling
: Fractional Actual Event
: Event Total Power Total
: 69.630853 0.846 70
|
n**********e
发帖数: 18 | 6 This one doesn't work
the code I used
proc power;
twosamplesurvival test=logrank
eventstotal=.
groupsurvexphazards = (0.001684 0.000838)
accrualtime = 0
followuptime = 12
power = 0.80
;
run;
the
could
Actual
【在 y*****w 的大作中提到】
: Thank you neardistance! If npergroup= and eventstotal= cannot appear at the
: same time, could you try one more time using the following code? If it could
: not run, then add "accrualtime=0". If you have already run the code like
: that, then stop here. I am asking this because there is no "Computing Actual
: Event Total" in your output, which is also what I would like to know.
: Thanks!
: proc power;
: twosamplesurvival test=logrank
: groupsurvexphazards = (0.001684 0.000838)
: followuptime = 12
|
y*****w
发帖数: 1350 | 7 Thank you!!! This gives me all I need to know. I am just a little surprised
that SAS output has no "Computing Actual Events Total", but just "Computing
Ceiling Events Total".
【在 n**********e 的大作中提到】
: This one doesn't work
: the code I used
: proc power;
: twosamplesurvival test=logrank
: eventstotal=.
: groupsurvexphazards = (0.001684 0.000838)
: accrualtime = 0
: followuptime = 12
: power = 0.80
: ;
|
