- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Hello,
I need to find the time period (3 month increments; Jan-Mar, Feb-Apr, etc.) where the majority of my data points (rates) are. The dataset extends 2 years. Would it be best to create a graph/visualization such as a scatterplot? Or use a proc freq statement on the dates? Any help would be greatly appreciated.
Accepted Solutions
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
If you have a date associated with the measures then 1) make sure it is a SAS date value (NOT character or numeric like 20180120 that only looks date-like) and 2) use a format like YYQ6. for the variable.
Then procs like Freq will group the variables into calendar quarters and display the value like 2017Q1 (first quarter of 2017)
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
"majority of data points" might mean something different to me than it means to you
Nevertheless, I can think of two approaches
- Histograms (bar charts) of the data
- Run PROC UNIVARIATE and use the MODE that is computed (MODE is the value occurs the most)
Paige Miller
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Thank you!
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
If you have a date associated with the measures then 1) make sure it is a SAS date value (NOT character or numeric like 20180120 that only looks date-like) and 2) use a format like YYQ6. for the variable.
Then procs like Freq will group the variables into calendar quarters and display the value like 2017Q1 (first quarter of 2017)
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Thank you. The proc freq did the trick. I did a cross-tabulation, so I could see the frequency of the ID's by date and quarter.