Hi There
how can I find SD in factorial experiment design . A, B, C ,D and E are factors and rep. is three replicate . data1, 2, 3 are measurements for example. what is the code for getting standard deviation in A*B, A*B*C, or C*D or etc.?
A B C D E rep. Data1 Data2 Data3
1 1 1 1 1 1 41 15 70
1 1 1 1 1 2 40 16 70
1 1 1 1 1 3 50 17 55
1 1 1 1 0 1 60 9 20
1 1 1 1 0 2 42 15 39
1 1 1 1 0 3 46 16 41
1 1 1 0 1 1 73 9 28
1 1 1 0 1 2 60 7 34
1 1 1 0 1 3 66 11 31
1 1 1 0 0 1 33 10 70
1 1 1 0 0 2 42 10 65
1 1 1 0 0 3 37 10 60
1 1 0 1 1 1 48 6 48
1 1 0 1 1 2 47 6 48
1 1 0 1 1 3 49 8 49
1 1 0 1 0 1 32 5 32
1 1 0 1 0 2 34 7 38
1 1 0 1 0 3 33 6 35
1 1 0 0 1 1 34 5 41
1 1 0 0 1 2 46 5 43
1 1 0 0 1 3 57 5 40
1 1 0 0 0 1 51 11 41
1 1 0 0 0 2 64 10 47
1 1 0 0 0 3 40 10 41
1 0 1 1 1 1 75 8 28
1 0 1 1 1 2 76 8 25
1 0 1 1 1 3 77 8 25
1 0 1 1 0 1 88 10 26
1 0 1 1 0 2 90 13 20
1 0 1 1 0 3 92 13 16
1 0 1 0 1 1 89 5 19
1 0 1 0 1 2 83 5 17
1 0 1 0 1 3 85 5 15
1 0 1 0 0 1 88 12 20
1 0 1 0 0 2 84 13 19
1 0 1 0 0 3 80 14 17
1 0 0 1 1 1 73 5 39
1 0 0 1 1 2 67 6 38
1 0 0 1 1 3 70 4 41
1 0 0 1 0 1 60 7 33
1 0 0 1 0 2 49 4 32
1 0 0 1 0 3 77 6 35
1 0 0 0 1 1 68 4 29
1 0 0 0 1 2 55 6 37
1 0 0 0 1 3 60 5 29
1 0 0 0 0 1 72 9 42
1 0 0 0 0 2 63 4 46
1 0 0 0 0 3 73 6 50
0 1 1 1 1 1 41 18 65
0 1 1 1 1 2 38 15 8
0 1 1 1 1 3 44 13 37
0 1 1 1 0 1 54 8 58
0 1 1 1 0 2 63 7 55
0 1 1 1 0 3 48 7 52
0 1 1 0 1 1 76 10 61
0 1 1 0 1 2 75 7 56
0 1 1 0 1 3 78 15 66
0 1 1 0 0 1 59 22 60
0 1 1 0 0 2 87 24 66
0 1 1 0 0 3 72 23 55
0 1 0 1 1 1 64 13 45
0 1 0 1 1 2 51 10 49
0 1 0 1 1 3 51 8 53
0 1 0 1 0 1 60 11 44
0 1 0 1 0 2 63 14 51
0 1 0 1 0 3 61 13 33
0 1 0 0 1 1 56 5 40
0 1 0 0 1 2 55 6 41
0 1 0 0 1 3 54 7 42
0 1 0 0 0 1 63 13 37
0 1 0 0 0 2 54 8 49
0 1 0 0 0 3 59 16 61
0 0 1 1 1 1 105 9 22
0 0 1 1 1 2 107 8 14
0 0 1 1 1 3 109 7 18
0 0 1 1 0 1 103 7 22
0 0 1 1 0 2 104 9 14
0 0 1 1 0 3 102 8 18
0 0 1 0 1 1 102 9 31
0 0 1 0 1 2 96 12 30
0 0 1 0 1 3 100 10 35
0 0 1 0 0 1 96 19 40
0 0 1 0 0 2 96 17 43
0 0 1 0 0 3 96 21 37
0 0 0 1 1 1 92 7 31
0 0 0 1 1 2 73 5 33
0 0 0 1 1 3 80 9 30
0 0 0 1 0 1 92 4 31
0 0 0 1 0 2 78 8 33
0 0 0 1 0 3 66 6 36
0 0 0 0 1 1 78 4 77
0 0 0 0 1 2 105 8 77
0 0 0 0 1 3 92 13 78
0 0 0 0 0 1 100 7 57
0 0 0 0 0 2 100 16 40
0 0 0 0 0 3 120 15 30
A simple standard deviation within combinations of the variables A through E can be accomplished with Proc Means or Summary;
proc means data=have stddev;
class A B C D E;
var data1 data2 data3 ;
run;
The columns A through E with 1 in the value are the variable combination used for the calculation
Or
proc summary data=have;
class A B C D E;
Var data1 data2 data3;
output out=want stddev=;
run;
which I usually find it a tad easier to read the output dataset than the way MEANS prints the data.
We don't normally speak of the standard deviation of the raw data within certain groupings with respect to a factorial design.
Normally we speak of the standard deviation of the parameter estimates, or of the residuals. Is that what you want?
Don't miss out on SAS Innovate - Register now for the FREE Livestream!
Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.