I've tried doing this in so many dirty ways, I figured it's time to ask for help.   I have 375 x variables that go x1-x375 and and one trial variable that counts up 1-180 in a single column. Each trial can have a different number of x's, anywhere from x1-x25 as the shortest all the way to x1-x375 as the longest. Yet every trial has x1-x375. The x's that don't have data are missing or a .   Trial X1 X2 X3 X4 X5 . . . . . . X374 X375   1    22  25 46 48  79                 .       .   I'm just trying to concatenate all the X's together. I tried:   x_cat = catx(",", of x:);   And it keeps returning an error "Invalid numeric data". I even tried filling in every . with another value then converting it to a character and I still get the same issue. Any help would be very much appreciated.     ... View more

I saw some previous solutions that don't quite fit, so I though I'd try to see if I could get some help.    I have two datasets. One dataset (time.dat) is very long (millions of rows) and contains three variables: subject, column, and time_n in this format:   subj col   time_n 2      1    1564623266038 2      2    1564623266044 2      3    1564623266058 2      4    1564623266073 2      5    1564623266090 2      6    1564623266106 . . .   Each subject has 1000's of columns that each have a unique time_n. There are also many subjects.   The second set (onloadtime.dat) has all of the same subjects and three variables, subject, trial, and onloadtime_n in this format:   subj trial onloadtime_n 2      1    1564623265779 2      2    1564623272356 2      3    1564623275373 2      4    1564623278728 2      5    1564623282137   Each subject only has 180 onloadtime's. What I need to do is find the closest time_n to each onloadtime_n and to indicate which trial corresponds to that time. Such that the resultant data looks like:   subj col    time_n                 onloadtime_n       trial 2      1   1564623266038    1564623265779     1   2      2   1564623266044 2      3   1564623266058 2      4   1564623266073 2      5   1564623266090 2      6   1564623266106 . . . 2    450  1564623272324    1564623272356   2   And to repeat this per subject. Essentially I need to know what trial in the shorter set (onloadtime) corresponds to what closest column in the long set (time), and to the indicate what time trial that actually is. I hope that makes sense. I'd love any support and would be happy to answer any questions.  ... View more

I'm trying to import multiple variables are each strings and can be more than 32767 a piece. I know the limit for a string variable is 32767, but I'm hoping to import such that if a variable reaches the max it spills over into another variable. I saw a partial fix in another post, but I can't get it to work. For trying it out, I limited the dataset to only 1 variable. To make it easier I also took the header row out o the dataset. This variable is just called Y. The current form is tab-delimited, but I've also unsuccessfully tried other formats.   Below I tried to import the Y variable as a possible array with a first variable Ypart1 up to 32767 and let it spill over into a Ypart2 if needed. As it is, it just fills up the first Ypart1 to the limit, makes a Ypart2, but no data appears there. I've attached the limited datafile. I can do this per variable, if necessary. Obviously it'd be best to do it all at once. Thanks for any guidance!   filename y_var temp; data y_test; infile 'C:\Users\dsm6713\Google Drive\Academic\y_var.tsv' truncover; array Ypart[2] $32767.; input (Ypart[*])($char32767.); run;   ... View more