Hi everybody, I have a dataset that I need to transform in a number of ways, and I'm very much stuck on how to do it. If anybody can help though, it would be beyond awesome. I have an example of what I have and what I would like to transform it into: Have: ID     QSN          VALUE 1      Prep            Zov 1      Prep            Pan 1      Prep            Dol 1      Strength      50 1      Strength      50 1      Strength      100 2      Prep            Zov 2      Prep            Zov 2      Prep            Zov 2      Strength      55 2      Strength      50 2      Strength      60 Want: ID      Zov      Zov_Strength      Pan      Pan_Strength      Dol      Dol_Strength 1         1          50                      1           50                      1           100 2         3          60                      0           .                         0             . To (try to)  put it in words: 1) When the QSN variable takes on the value "Prep", i need the value of the value variable to be a variable in the new dataset (example "Zov"). 2) The value of this variable, should be the number of times it showed up under the ID. 3) When the QSN variable takes on the value "Strength", it corresponds to a value of Prep. 4) If there are more than one prep (example "Zov") and the strength variables are diffierent, the correct one should be the highest. 5) The data in the original data set is ordered so that the "prep"s come first, then the corresponding "Strength"s follow. I hope this makes sense and if anybody can help me out, I would be incredibly grateful. Best regards Thomas ... View more

Hi everybody, I have data on dates and costs of visits, and I need to add them up over individuals seperated into months. Dummy data is something like this: Have: Id            Date            Cost       Visit_id 100       10sep2005     1000       98 105       20sep2005     3000       103 105       22sep2005     1000       105 200       01oct2005      2000       123 200       03nov2005     1500        201 200       03nov2005      500         202 Want: Id          Month          Visits          Cost 100       sep2005         1             1000 105       sep2005         2             4000 200       oct2005          1             2000 200       nov2005          2             2000 I've tried using the ".first" and ".last" properties when creating a loop, but I can't figure out exactly how to do it. Any help would be greatly appreciated.          ... View more

Hi Everybody, I need help separating two variables on a monthly basis. Specifically, I have data on patients admitted to hospitals. The four variables are: d_in: The date the patient was admitted to the hospital d_out: The date the patient was released from the hospital t_stay: The total number of days the patient stayed at the hospital t_cost: The total cost of the stay. I need a way to get the t_stay and t_cost separated by month; for example, if a patient was admitted to the hospital from the 20 th of april to the 10 th of may and the total cost was $200, I would get the values stay_april=10, stay_may=10, cost_april=100 and cost_may=100. If the patient is admitted and released the same month, it is easy, since I can extract the month of the start and the month of the end, check with an if-statement if they are the same, and if they are, I have the values I need directly from the t_stay and t_cost variables. But I'm having trouble figuring out how to do this with patients whose stay span several months? I hope my question makes sense and I will be grateful for any help. ... View more