Rick Wicklin wrote: It's an intriguing idea, but if it were me I'd use the simpler NPAR1WAY procedure. To me, the simpler procedure has the advantage of more explicit assumtions and (often) more power than the general test. (And it's easier for me to understand what is being tested.) By way of an analogy, if I'm testing for the difference of means, I wouldn't use PROC GLM and the ESTIMATE statement when I could use PROC TTEST. I'm not sure that the stated criterion of "(often) more power" is a useful one. Especially when "(often)" is demonstrated by a single example. Do you have references regarding power? What "explicit assumtions (sic)" are present in NPAR1WAY that are not present for QUANTREG's test? ... View more

AllenMcDowell wrote: The mathematical details of the Wald test are provided at: http://support.sas.com/documentation/cdl/en/statug/65328/HTML/default/viewer.htm#statug_qreg_syntax11.htm You can also perform a likelihood ratio test by specifying the LR option in the TEST statement; the mathematical details are provided on this same page. With all due respect, there are no mathematical details of the Wald test at this link. ... View more

Thanks, Allen. That looks like a reasonable approach. Do you have any references on the Wald test? I don't really find much in SAS docs for PROC QUANTREG about the Wald test. ... View more

Arthur Tabachneck wrote: "When you perform these tests, your data should consist of pairs of measurements for a random sample from a single population." Okay, that does not describe my situation. ... View more

Arthur Tabachneck wrote: You can also use proc freq.  Take a look at http://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_intronpar_sect006.htm I don't see the word "median" anywhere in the PROC FREQ documentation. ... View more

It was always my understanding that the median test in PROC NPAR1WAY was a non-parametric test of means, not of medians. ... View more

Nice job with the SQL code, PGStats! However, "fairly simple" to you is not fairly simple to the many of us. ... View more

Actually, my suggestion to expand the dataset so it contains appropriate rows with 0, can be done via a combination of DATA steps and PROC SQL. The computation of the moving average could be done via LAG commands in a DATA STEP, but it is easier to use PROC MACONTROL (if you have SAS/QC) ... View more

LinusH wrote: One "easy" way to do it is to create a multilabel format that groups your moving quarters. Then use that format together with PROC MEANS, using the MEAN statistic. http://support.sas.com/documentation/cdl/en/proc/65145/HTML/default/viewer.htm#n13dhme6o4ut3en1u8e58r16b3zf.htm Unless I am missing something, this seems like an awfully work-intensive way to get moving averages, despite your use of the word "easy". Don't you have to pre-define the quarters as overlapping formats in a PROC FORMAT to make this work? Which implies in advance that you have to know exactly what quarters are in the database, and code that up into PROC FORMAT. If you have 50 quarters, that's a lot of PROC FORMAT coding, and the slightest error in there invalidates the results. Am I right, or did I miss something? ... View more

I'm not going to write code for you, but your table needs to be augmented with rows that contain 0 for the quarters where the quantity is zero. Otherwise, you will never get the proper moving averages. PROC MACONTROL will do the calculations for you, even allows for BY groups, but I think first you must augment the table with zero values. If you don't want the control charts that PROC MACONTROL produces, you can ignore/eliminate those and use the output data set (OUTHISTORY=) which contains your moving averages. ... View more