I am not sure exactly what you want but you could make a month-only variable by adding a new column with the simple formula
Month(:Date) //assuming Date is the name of your date column
This will yield a numeric column with numbers 1-12. These can be recoded (use e.g. Recode in the Cols menu) to Jan-Dec if you prefer that. Use this column with Summary or Tabulate to get monthly statistics.
However, if you want summarize by both month and year then you could also make a Year(:Date) column and then group after both columns.
Alternatively make a a combined column with the formula (choose format m-y to avoid showing the date)
Date DMY( 1, Month( :Date ), Year( :Date ) )
Or by concatenating the month and year columns I described above using the column formula
Char( :Month ) || "-" || Char( :Year )
The last example gives a nominal month-year though, which may or may not be desired.
I think I just came up with a more elegant solution in a similar recent thread. I thought the problem seemed familiar, that I've been digging in to this not long ago. How ephemeral memory is! Glad you've found a solution.
There is a Week of Year function which assumes weeks begin on Sunday so you may get a week fragment at the beginning and another one (week 53) at the end of a year.
There is a Day of Year function which can be used to calculate a different Week of Year value which starts a full 7 day week on the first day of the year. There will be a 1 or 2 day week 53 fragment at the end of the year. That formula is:
Floor( (Day Of Year( :Date ) - 1) / 7 ) + 1
One way or the other you will have a new column of weeks of the year.
Well you can use the Year function to get a column containing the year. Then you can model by Year and Week. Or you can make a value like your 200940 by creating a new column whose values are Year*100 + Week if that's what you want. There is no JMP format which displays a date as YYYYWW.
i am doing something similar to that described by MS (actually working out average yields by month) and haven't yet found a nicer way.
unfortunately, if i want to plot a bar chart with months as the x-axis and there are months with no data, then that month doesn't get plotted at all, so to make sure there's an empty month in my chart i have to create dummy empty rows, and i'm sure there should be a better way than that!