Hi, I have a question about how to put duplicates into different groups based on the original order? I have a data like below:   ID condition 1 A 1 A 2 B 2 B 3 A 3 A 3 B 3 B 4 B 4 B 4 A 4 A 5 A 5 A 5 B 5 B 5 A 5 A The data is like above, basically, there are five situations: 1. all As; 2, all Bs; 3, A first and then B; 4, B first and then A; 5, A first and then B, and then A. I want to keep the original order, which means I can't use proc sort, because after I use proc sort, AABBAA or BBAA will become AAAABB and AABB. What I want to do is to put ID with all As in group 1, put ID with all Bs in group 2, and put ID with A first and then B in to group 3. So, for the rest of patterns like B first and then A, or A first and then B, and then A can be removed. I am wondering if there is a way to do it? Hopefully that I express myself clearly. Thank you! ... View more

Hi, the situation is little bit complicated, so the title might not describe it well. I have two dataset with same subjects, one dataset is cleaned with one row and one unique subject ID and group variables, like below: ID Group 1 Group 2 Group 3 1 1 0 0 2 0 1 0 3 0 0 1 4 0 0 1 and other one is not cleaned with multiple same IDs with a condition variable, like below: ID Condition 1 Long strings example 1 2 Short string example 2 3 Long strings example 1 3 Long strings example 1 3 Short strings example 2 4 Long strings example 1 4 Short strings example 2 Original condition contains really long strings so I used "long strings example 1" and "short string example  2" as examples. Generally, if the ID is in the group 1, this subject is "long strings example 1" only; if the ID in the group 2, this subject is "short strings example 2" only; if the ID in the group 3, this subject has both "long strings example 1" and "short strings example 2" several times.    Here is what I want to do, I only want two groups: group 1 and group 2. Group 3 here just is used to mark this ID has both "long" and "short" several times(but I don't know how many times for each "long" and "short"). So if the ID has 2 or more "long strings example 1", I want to divide it into group 1. Here, because ID 3 has more than 2 "long"s, the ID 3 should be divided into group 1. And if ID has one "short strings example 2" and the "long strings example 1" is less than 2, then this ID is belongs to group 2. Here, because ID 4 has only 1 "long", and one "short", it belongs to group 2. So, I need to work on the two datasets together.   Eventually, the dataset should look like belows: ID Group 1 Group 2 Group 3 Final group 1 1 0 0 1 2 0 1 0 0 3 0 0 1 1 4 0 0 1 0   In the final group, 1 is group 1, and 0 is group 2.   Could anyone helps me with it? Hopefully, I made myself clearly. Thank you! ... View more

Hi, I have a data pretty much like below: Type Year A B C 1 2011 0 1 0 1 2011 1 0 1 1 2012 1 1 1 1 2012 0 1 0 2 2011 0 0 0 2 2011 1 0 1 2 2012 1 0 1 2 2012 1 1 1 And I want to compute the percentage for A, B, C by each group Type and Year, the result is pretty much like below: Type Year A percent B percent C percent 1 2011 50% 50% 50% 1 2012 50% 100% 50% 2 2011 50% 0% 50% 2 2012 100% 50% 100% Do you know how to do it? When I used the below code, the result was different from what I expected, it had lots of years for each type. proc sql; create table rate as select type, year, A, B, C, sum(A)/count(A) as pctA, sum(B)/count(B) as pctB, sum(C)/count(C) as pctC from (select type, year, A, B, C from one group by type, year) group by type, year; quit; Could anyone help me with it? Thank you! ... View more

Hi, I tried to import excel file into SAS, and it has time. In the excel, the time format is like: 12:33:00 AM, even though I tried to change the format and remove the 'AM', but in the formula bar, the 'AM' is still there, only in the cell the 'AM' was gone. So I imported the file into SAS using "proc import", the time from 12:33:00 AM changed into 2.291666666666666669E-2 and it's character. I tried to convert it into numeric by using 'input(time,anydtdte32.)' and 'input(time, mdyampm22.)', they didn't work. Only 'input(cats(time,"00"),hhmmss.)' worked for some of the values, but didn't work for the values like "1.25E-3" (the ones without long repeating decimals like 2.29166666666666666E-3).    I am wondering if there is a way to remove AM/PM in the excel completely not even on the formula bar? Or in the SAS, after the import, how to convert it into numeric first and then into seconds? Or convert them into seconds directly? Could anyone help me please? Thank you! ... View more