Period 1 2 3 4 5 6 ... ... ... ... ... ... ... 201401 22223 30303 40000 80000 123444 144000 201402 22334 40002 100202 120000 132000 . 201403 129934 150000 123444 130000 . . 201404 54349 75000 123444 . . . 201501 99900 134423 . . . . 201502 102030 . . . . . I have created triangles of data in SAS. When I say triangles, the cells off the triangle are left blank. The actual triangle is much bigger than this one and I'm trying to write the code to sum over the bottom nonblank cells of each column. I'm not sure if this is possible but I would think it would involve proc summary. Are there any functions that can achieve this? Thank you   ... View more

I'm getting an error when I run this code: %let cutoff = 201506; %let averageperiod = 8; %let startqtr = %SYSFUNC(intnx('qtr', %input(%put(&cutoff*100+ 01,8.),yymmdd8.), -&averageperiod, 'same')); %put &startqtr; the error says... ERROR: Open code statement recursion detected. 885  %let averageperiod = 8; 886  %let startqtr = %SYSFUNC(intnx('qtr', %input(%put(&cutoff*100+ 886! 01,8.),yymmdd8.), -8, 'same')); ERROR: Open code statement recursion detected. 887  %put &startqtr; ERROR: Open code statement recursion detected. ... View more

I have a filename defined in my dataset:   filename cm  'k:\may 2015\cm.txt'; and in another program the month and year are defined as: %let month = may; %let year = 2015; when I used the code filename cm  'k:\&month. &year.\cm.txt'; it doesn't work ... View more

I am trying to use sas code to create two extra columns on my dataset as follows: Her is my sample dataset: Veh No Prem start end 1 500 1/1/10 1/3/10 1 0 2/3/10 1/6/10 1 0 2/6/10 30/9/10 1 0 1/10/10 31/12/10 2 600 1/1/09 1/3/09 2 700 1/1/09 1/3/09 2 0 2/3/09 20/9/09 2 0 21/9/09 30/9/09 I want to create three extra fields as follows: prem1 start1 end1 500 1/1/10 31/12/10 500 1/1/10 31/12/10 500 1/1/10 31/12/10 500 1/1/10 31/12/10 600 1/1/09 1/3/09 700 1/1/09 30/09/09 700 1/1/09 30/09/09 700 1/1/09 30/09/09 So for each vehicle number where a non-zero premium is followed by a certain number of records with a zero premium I want to retain the premium and start date fields for all the 0 premium records and retrofit the last end date of each sequence of (premium followed by 0 premium for each vehicle) What I'm trying to do is allocate the premium across all the exposure records in proportion to the length of exposure ... View more

I want to use a formatting table to map months to quarters. I've created a format table below called "quart". I've included a sample dataset with months but I've included two months outside the range of the format table. However, when I use the code check=put(month,quart.); I still get values for the months outside the range of the format table, specifically, 1 for month 13 and 2 for month 21. Is there a way to change the code so that the code is only executed where the months are in the range of the format table. Something like if check=put(month,quart.) ne "" then do; Here is the sample code data quart; input start $2. label $1.; datalines; 011 021 031 042 052 062 073 083 093 104 114 124 ; run; data quart; set quart; fmtname = "quart"; type = "c"; run; proc format cntlin = quart fmtlib; run; data dataset; input month $2.; datalines; 01 02 13 21 ; run; data dataset; set dataset; check=put(month,quart.); run; ... View more

My understanding is that this function returns a position of the first letter of the alphabet in a string. Is there a similar function that include all non-numeric symbols. The code I'm using is if anyalpha(field1) eq 0 then field2 = field1; else field2 = 0; But I'm not sure what function I should use if there is "/", "-", etc. ... View more

I have a field Header1 ($20) of the form: Header 1 485 1 AB27834 134 A/DD123 678423167 AB1266 12 and where the cell is a string where the characters are number I want to convert to the format; 0000000001 where Header1 is 1 0000000134 where Header1 is 134 and so on. for the cells where the string contains a mix of letters and number I want to leave them as they are. I tried using this code: Header1new = put(input(header1,20.),z10.); This converts the numbers the way I want but it produces loads of errors for the non-numbers. Is there code that allows for this i.e. if the string header 1 contains at least one character then do leave as is. Thanks ... View more