turn on suggestions

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

Showing results for

- JMP User Community
- :
- Discussions
- :
- Re: Clustering Analysis?

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
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Apr 6, 2017 8:47 PM
(1552 views)

Can I use clustering analysis with continuous variable target? I am using SAS jmp. I tried using continuous variable for hierarchical clustering but the answer returned an error. What am I missing?

Attached is the data set.

Salesprice is my continuous target variable

1 ACCEPTED SOLUTION

Accepted Solutions

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Apr 7, 2017 5:37 AM
(3053 views)

Solution

- Select
**Analyze**>**Clustering**>**Hierarchical Cluster**. - Select
**SalesPrice**. - Click
**Y, Columns**. - Click
**OK**.

I did not encounter any errors up to this point.

This unsupervised method found eight clusters, each of which with two sub-clusters, as indicated in the dendrogram or the constellation plot.

Learn it once, use it forever!

2 REPLIES

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Apr 7, 2017 5:37 AM
(3054 views)

- Select
**Analyze**>**Clustering**>**Hierarchical Cluster**. - Select
**SalesPrice**. - Click
**Y, Columns**. - Click
**OK**.

I did not encounter any errors up to this point.

This unsupervised method found eight clusters, each of which with two sub-clusters, as indicated in the dendrogram or the constellation plot.

Learn it once, use it forever!

Highlighted
I don't understand this solution. What is the sense of clustering on a single target variable? I thought the whole idea of clustering was to find clusters of observations that are "similar" and then perhaps might related to a target variable. How is clustering on a single target variable any different from looking at a distribution of that variable and seeing if there are any groups? And, from your solution, what are the X,Y coordinates of the constellation (when there is a single variable being clustered)?

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Apr 20, 2017 10:10 AM
(1415 views)