Suppose one is attempting to run regression on the following two economic indicators:   Employment and Business Cycle   (please see graphic below)     Linear Regression won't work here because level of Business Cycle (say, from 0 to 100) includes both 'coming down' and 'going up.'   The regression coefficient for Business Cycle will always be not significant because Business Cycle values will include offsetting 'going up' and 'coming down' values.   Yet one knows that Business Cycle level is indeed important.   One would additionally need to include a second variable -- whether the cycle is 'going up' or 'coming down.'  Change from Previous Period sounds appropriate.   Thus, Multiple Regression might be usable.   I'm curious on your take.  How can one best go about working with this oscillator dilemma?   Thanks!   Nicholas Kormanik     ... View more

Please view the following page, at the very bottom:   http://support.sas.com/documentation/cdl/en/statug/68162/HTML/default/viewer.htm#statug_glmselect_examples01.htm   Attached is a screenshot of the relevant section.     I'm trying to interpret this.  Seems to me that the 'model' presented is not so good.  These are standardized coefficients.  The largest, by far, is the Intercept.  All the rest are relatively very small.   Would you interpret the results this way?  What am I missing?   Thanks!   Nicholas Kormanik     ... View more

Suppose we have a number of column vectors of data, each column representing a case.  Suppose 10 such columns.   Rows represent characteristics.  Suppose 50 such characteristics.   Some cells in the matrix are empty -- missing data of Case Xi on Characteristic Yj.   The question is, How can one describe the similarities among the given cases?   A parallel example, perhaps:  Images of people.  Two eyes, a nose.  But, some short hair, some long, so not similar on that measure.   So, what are the similarities among the cases?   All suggestions and thoughts appreciated.   Thanks.   Nicholas Kormanik     ... View more

Windows 10 Pro, 64-bit. SAS 9.4 TS1M3, 64-bit.   For years drag-and-drop has worked perfectly -- SAS programs in text files on desktop clicked on and dragged onto SAS window, log or editor. No fuss, no muss.   Suddenly, maybe because of a Windows 'update,' the ill-behavior of some other installed program, or because of malware, the wonderful drag-and-drop for SAS no longer works.   When the dragged file gets over the SAS window, a black circle with black line diagonally through circle appears. No 'drop' is possible.   Fortunately, copy-contents-first and then paste-to-editor still works.   Does anyone have any idea how to possibly correct this awful problem??   Some registry key? Some group policy setting? Some other program I need to uninstall?   Thoughts greatly appreciated.   Nicholas Kormanik     ... View more

I just finished doing some slicing-and-dicing of a character field, using SAS built-in functions.  Involved many steps and seemed a lot more complicated than I would have guessed.   Just curious, do most advanced users here graduate to another approach to character field manipulation?   I see that SAS also allows using RegEx.  Would RegEx be a better approach to many character field manipulations?   Here is an example of one manipulation I was attempting:   Have: rg1_23456a_7   Want: 34a Thanks much.   Nicholas Kormanik     ... View more