Thanks for providing the sample. Will you be reading in many of these, or just the one?   You can certainly do this with a data step, though it will probably be best to read in line-by-line. You can use the scan function, for example, to parse the input strings. The DATA step will allow you to output to different data sets, so you could output state names to a temporary set, the numeric values to another, and append them. This is difficult because your data doesn't follow typical CSV patterns.   However, it may be easier to reformat the set, especially if you are only reading in the one set. I copied and pasted the data you posted into Excel. I went to Data >> Text to Columns and selected the options of delimeter, comma. Then I formatted it like this: State Note Total Taxes Year Quarter Revision U.S. Total Excludes D.C. 133987401 2000 1 1 Alabama 1554648 2000 1 1 Alaska 204757 2000 1 1 Arizona Revised 1922171 2000 1 1 Arkansas Revised 1135907 2000 1 1 California 20206217 2000 1 1 Colorado 1614872 2000 1 1 Connecticut Revised 2219791 2000 1 1 Delaware Revised 533972 2000 1 1 Florida Revised 6664802 2000 1 1 Georgia 3295666 2000 1 1 Hawaii 860653 2000 1 1 Idaho Revised 539018 2000 1 1 Illinois Revised 6142447 2000 1 1 Indiana Revised 2492170 2000 1 1 Iowa 1363449 2000 1 1 Kansas 1104084 2000 1 1 Kentucky Revised 1726561 2000 1 1 Louisiana Revised 1494620 2000 1 1 Maine 583932 2000 1 1 Maryland 2267078 2000 1 1 Massachusetts 4055999 2000 1 1 Michigan 6827675 2000 1 1 Minnesota 3719413 2000 1 1 Mississippi 930371 2000 1 1 Missouri Revised 1879749 2000 1 1 Montana Est. 317540 2000 1 1 Nebraska 722445 2000 1 1 Nevada Revised 626221 2000 1 1 New Hampshire 287934 2000 1 1 New Jersey 4029776 2000 1 1 New Mexico Revised 831934 2000 1 1 New York 11762936 2000 1 1 North Carolina 3695397 2000 1 1 North Dakota 274030 2000 1 1 Ohio 5757836 2000 1 1 Oklahoma Revised 1354611 2000 1 1 Oregon 1307216 2000 1 1 Pennsylvania 6049117 2000 1 1 Rhode Island 508208 2000 1 1 South Carolina 1325561 2000 1 1 South Dakota Revised 190918 2000 1 1 Tennessee 1925580 2000 1 1 Texas 6503198 2000 1 1 Utah 867827 2000 1 1 Vermont Revised 341961 2000 1 1 Virginia 2786879 2000 1 1 Washington 3131704 2000 1 1 West Virginia 765184 2000 1 1 Wisconsin 3069464 2000 1 1 Wyoming 213902 2000 1 1 Wash. DC 628645 2000 1 1 The formatting looks off here, but you should be able to copy into Excel. Then PROC import is trivial. Or format it yourself to suit your purpose. ... View more

This depends on your data and how much you want SAS to do. For one, you can use DSD and MISSOVER if you use a DATA step. The DSD option will help if you have several commas in a row that signify missing values. This paper is a good intro to some options.   How does the code you have now work? If you still need help, can you post a small example of 0001.csv, including some of the "worst" examples you are seeing? Maybe including what looks wrong in the output would help. ... View more

with your updated requirement, is this what you want? note that the Y value when X is 3 will be .   data temp; input X Y; datalines; 1 70 1 . 1 . 1 . 1 80 1 . 1 . 2 60 2 . 3 . ; data temp2; set temp; retain last_non_missing; if (first.x) then last_non_missing = .; if (not missing(Y)) then last_non_missing = Y; by x; rename last_non_missing=Y; drop Y; run; ... View more

Remove the rename and drop statements if you want the new variable to be in its own column.   data temp; input X Y; datalines; 1 70 1 . 1 . 1 . 1 80 1 . 1 . ; data temp2; set temp; retain last_non_missing; if (not missing(Y)) then last_non_missing = Y; rename last_non_missing=Y; drop Y; run; ... View more

This Stack Overflow thread suggests using the following: filename sas2xls dde 'excel|system'; data _null_; file sas2xls; put '[open("C:\Users\Dmitry\Google Drive\SAS\test.xlsx")]'; put '[workbook.hide("Sheet1")]'; put '[error(false)]'; put '[save()]'; put '[file.close(false)]'; run; ... View more

Use the INTNX function.   For example:   intnx('month', yourVariable, -1, 'same'); will return the same day in the previous month (-1) Use 'b' or 'e' to return the beginning or end day of the previous month, respectively.   More information and examples. ... View more

This sounds like a great conference. For more information, or if you are unable to participate in the conference, two great resources are expert J. Scott Armstrong's list of Standards and Practices for Forecasting and this great paper reviewing hundreds of other works and a myriad of forecasting methods and techniques. ... View more

These look great! Thank you for sharing.   You should submit a pull request to Prism with these. They are big improvements upon what they already have, and other SAS+Prism users will benefit. ... View more

The link you posted mentions SAS support already. I'm hoping your .js is more complete than this one (origin?), though anything would be useful. ... View more

That would be fantastic! I would definitely use it. Is it in a GitHub repository? If it isn't to be made public, you can email it to me. ... View more

Well the SAS online trainings are great, and interactive! You could finish one or two in a week, if you have the time!   I hope Reeza's suggestion helps, and good luck with the project and paper. ... View more

Well, a benefit of writing code would be to give you the power and control to something that doesn't have an explicit option.   I'm not sure how you can do this via the graphical interface. ... View more