Pages: [1]
  Print  
Author Topic: Finding trends and predicting sales figures  (Read 617 times)
rabarbar
Newbie
*
Posts: 1


« on: September 23, 2013, 11:28:45 AM »

Hello everyone!  Smiley I couldn't find solution for my problem, so i decided to register here and ask you  Wink

I have a table with sales and time, like this:

datesales
2011/01500
2011/02600
2011/03700
2011/04800
2011/05900

I triend linear regression, after setting date as ID and sales as label, I have no other attributes... How can I handle this?

How can I predict sales in next months? Of course this dataset is very simple, just to understand the tool, later I would like to apply this to non-linear issues.

Thanks for your help!
« Last Edit: September 23, 2013, 12:35:52 PM by rabarbar » Logged
Marius
Administrator
Hero Member
*****
Posts: 1793



WWW
« Reply #1 on: October 02, 2013, 01:55:43 PM »

Do not set the date as id, but use it as input variable, since you want to predict sales over time.

It may be possible that the underlying Java will have problems with dates as regular attribute. In that case, use the Set Role operator to create an id (a running idx), set the role of id to regular and in this case the date column back to id.
Then you can use the linear regression to get the job done.

Best regards,
Marius
Logged

Please add [SOLVED] to the topic title when your problem has been solved! (do so by editing the first post in the thread and modifying the title)
Please click here before posting.
Pages: [1]
  Print  
 
Jump to: