Datasp, When I made the statement to include additional columns, I meant I may need to add other variables to my data, not just including more months and years. After I tested with the solution provided by BILLFISH, I realized that, as you said, the denominators vary.  I need a fixed window-size, such as only looking back at the previous four years from the current year or year in question.  So, 2015 needs to be compared to the average of 2014 to 2011, 2014 needs to be compared to 2013 to 2010, and so on. Yes, BILLFISH gave a solution to what I asked.  It is my fault that I didn't clarify enough in writing that I want a moving average calculation, but only have the average calculate for the prior four years from a given year.  I'm hoping based on the solutions I have been given that a line or two of code can be added or modified for them to work in the manner I intended. Thank you for following up and being patient with me on this. ... View more

I think you got it!  Thank you!!!  I'll let you know once I try it with my actual data.  Thanks again! ... View more

My table consists of Year, Month, Sales, and Running Sum (of Sales).  I want to calculate the moving average of the Running Sum (of Sales). From my sample data provided, I know the Running Sum (of Sales) through Jan 2013 is 150, Feb 2013 is 425, and March 2013 is 675.  How do I write the code in EG to calculate the moving average to get 138 for Jan, 413 for Feb, and 625 for March? From the research I have done, it appears I should use PROC EXPAND to write the code.  If there is another method I should use, please explain.  Down the road, I may need to include additional columns to my data and perform the calculations at that level of detail. ... View more

Reeza, I apologize for the confusion.  Here is what I am trying to calculate. Year     Month     Sales     Running Sum     Average 2011     Jan          100          100                    . 2011     Feb          250          350                    . 2011     March      200          550                    . 2012     Jan          175          175                  100 2012     Feb          300          475                  350 2012     March      225          700                  550 2013     Jan          150          150                  138  ((175+100) / 2) 2013     Feb          275          425                  413  ((475+350) / 2) 2013     March      250          675                  625  ((700+550) / 2) The Average column calculates the Running Sum through each month and divides that figure by the number of prior periods.  See the calculations in parentheses.  I'm actually trying to calculate a four year average, but I know I did not include enough data in my sample to calculate the full four years. I am able to use PROC EXPAND to calculate a moving average, but I only get the code to work when I had the Year and Sales (at cumulative total) columns in my dataset.  Sales were for the full year, not broken out by month.  For 2011, sales were 550.  For 2012, sales were 700. So, how do I calculate the moving average with more than two columns in my table? I hope this explanation helps answer your questions.  Thanks again. ... View more