turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- BI
- /
- Visual Analytics
- /
- What is the best way to build a statistic model to...

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

07-12-2017 02:25 PM

Hello,

I need to predict the values for Y (dependent variable). Y is numeric but in our case it is not a continuous value but rather a discrete and categorical value with order. The X (independent variables) are categorical values as well.

I need to predict Y over time, e.g. over the next six months. What is the best way to build this model in SAS Visual Statistics? I was trying to build a time series line chart with the forecast. However, the Y values are aggregated because there are multiple values on a given day. I don't need the aggregated (either sum or average) values.

Thanks,

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

Posted in reply to poppies

07-12-2017 03:24 PM

If you aren't aggregating how are you differentiating between the different values on the same day?

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Reeza

07-12-2017 03:26 PM

I guess that's my question...since aggregating the values probably doesn't make any practical sense in our case.

Also, given the data types, is it appropriate to use the series line chart to do the forecasting?

Also, given the data types, is it appropriate to use the series line chart to do the forecasting?