2 Replies Latest reply: Sep 12, 2012 1:00 PM by mjjouana RSS

    How to calculate moving averages with a big database requiring filtering

      Hello,

       

      I have a huge database containing customers information in which some fields are corrupted with irrelevant data and I need to produce a chart plotting  subscription dates on the x-axis and the time required to respond to customers orders. I aslo need to overlap a moving average on the chart.

       

      Given the dimensions are: date of subscription, date of response, number of days required to respond to order,

      how can I:

      - filter the date of response field which contains values other than dates and previous dates

      - have a chart displaying all the dates of subscription on the x-axis (I have around 14000 entries) and the number of days on the y-axis

      - overlap a moving average for the number of days ( each day, each 7 days or each month...)

       

      Thank you all for your help & Regards

       

      Marie Joelle

        • Re: How to calculate moving averages with a big database requiring filtering
          Wojciech Parzyszek

          no big deal,

          send sample qvw file (with data)

            • Re: How to calculate moving averages with a big database requiring filtering

              Sorry for the very late reply. I've attached the samples to the original post. Here's the situation:

               

              I have huge files from different sources with common key fields (Client number and product number).

              In one file, I have the date of subscription (dates of order placement), in the other file, date of response (delivery date).

               

              The file containing date of response is corrupted with different types of data (other than dates). Also, for some entries, the date of response is older the date of subscription. Several entries have the same date of subscription and of response but are for different clients and one client can have different types of products.

               

              I have to calculate the difference between the two dates but also do a moving average for say every 3 days and display the results in a chart. (However, the number of entries, having similar dates but concerning different clients, is quite random from one month to another)

               

              Thank you so much for your help,

              Regards