comp.soft-sys.sas - The SAS statistics package.
I am using PROC DISCRIM to calculate the accuracy of continuos variables (speed, degree flexibility...) in telling appart cows with ulcers in their hooves. We checked the hooves of the cows and measured the predictors the same day. I ran the code: Proc DISCRIM; class ulcer; var speed; run; where ulcer is a dychotomous variable (yes or not). I saw the output gives you 2*2 tables with specificity and sensitivity, that it is what I want. However, is there any place in the output where the cut-off point used for that is specified?
I have a problem when using the %ds2csv macro. when making the csv file lines get cut off at 256 characters. How can this be solved???
Do you know a way to create a variable that flags the quartile (or decile) a case falls into for a given variable. For example, the lowest quartile of income would be '1', and 2nd quartile would be '2' etc etc.
For TAGSETS.EXCELXP I'm using embedded_titles='yes' autofit_height='yes' fittopage='yes' pages_fitwidth='1' pages_fitheight='10' I created a title using title1 h=15pt 'Test' However, the word "Test" doesn't fit in the cell (of the 1st row). The top part of the letters are cut-off. Is there any way to increase the height of the 1st row in the Excel worksheet? Thank you.